1
Cell-substrate adhesion drives Scar/WAVE 1
activation and phosphorylation, which controls 2
pseudopod lifetime 3
4
Shashi Prakash Singh+, Peter A. Thomason+, Sergio Lilla+, Matthias Schaks$, 5
Qing Tang%, Bruce L. Goode%, Laura M. Machesky+*, Klemens Rottner$ and 6
Robert H. Insall+* 7
8
+CRUK Beatson Institute, Switchback Road, Glasgow G61 1BD, UK; *Institute 9
of Cancer Sciences, University of Glasgow, Switchback Road, Glasgow G61 10
1BD, UK; $Zoological Institute, Technische Universität Braunschweig, 11
Spielmannstrasse 7, 38106 Braunschweig, Germany & Cell Biology, Helmholtz 12
Centre for Infection Research, Inhoffenstrasse 7, 38124 Braunschweig, 13
Germany; %Brandeis University, 415 South Street, Waltham, MA 02453, USA. 14
15
Correspondence to Robert Insall 16
([email protected]; tel. +44/0 141 330 4005) 17
18
2
Abstract 19
The Scar/WAVE complex is the principal catalyst of pseudopod and 20
lamellipod formation. Here we show that Scar/WAVE’s proline-rich domain 21
is polyphosphorylated after the complex is activated. Treatments that stop 22
activation block phosphorylation in both Dictyostelium and mammalian cells. 23
This implies that phosphorylation modulates pseudopods after they have 24
been formed, rather than controlling whether a protrusion is initiated. 25
Unexpectedly, activation-dependent phosphorylation is not promoted by 26
chemotactic signalling, or by signal-dependent kinases such as ERKs, but is 27
greatly stimulated by cell:substrate adhesion. Scar/WAVE that has been 28
mutated to be either unphosphorylatable or phosphomimetic is activated 29
normally, and rescues the phenotype of scar- cells, demonstrating that 30
phosphorylation is dispensible for activation and actin regulation. However, 31
pseudopods and patches of Scar/WAVE complex recruitment last 32
substantially longer in unphosphorylatable mutants, altering cell polarisation 33
and the efficiency of migration. We conclude that pseudopod engagement 34
with substratum is more important than extracellular signals at regulating 35
Scar/WAVE’s activity, and that phosphorylation acts as a timer, restricting 36
pseudopod lifetime by promoting Scar/WAVE turnover. 37
3
Introduction 38
Scar/WAVE is the dominant source of actin protrusions at the edge of 39
migrating cells. In particular, lamellipods (in mammalian cells cultured in 40
2D) and pseudopods (in cells in 3D environments, or cells such as amoebas) 41
are driven by Scar/WAVE recruiting the Arp2/3 complex, which in turn 42
promotes an increase in the number of polymerizing actin filaments and 43
growth of actin structures (1). It works as part of a large, five-membered 44
complex, whose members have multiple names (2); in this paper they will be 45
referred to as Nap1, PIR121, Scar, Abi and Brk1 in Dictyostelium, and Nap1, 46
PIR121, WAVE2, Abi2 and HSPC300 in mammals. 47
48
The principal known activator of Scar/WAVE complex activation is the small 49
GTPase Rac. Inactive Rac is GDP-bound, but on stimulation becomes 50
temporarily GTP-bound. The GTP-bound, but not the GDP-bound form 51
binds to the complex (3), in particular through the A-site of PIR121, which 52
includes a Rac-binding DUF1394 domain (4). Interaction with GTP-bound 53
Rac is essential for the complex to be able to function (3, 5). However, though 54
it is clear that Rac is essential, it is not the only regulator (1). Various 55
experiments have found that Rac activation occurs later than the onset of 56
actin-based protrusion (6), and signal-induced actin polymerization can occur 57
earlier than Rac activation (7). In Dictyostelium, Scar/WAVE behaviour is 58
much more locally variable than Rac activity (8), so the Rac cannot simply be 59
driving the changes in Scar/WAVE. To understand pseudopod dynamics, it 60
will therefore be vital to enumerate different modes of Scar/WAVE regulation. 61
One potential form of regulation – phosphorylation - has been described in a 62
number of papers, and is also found in untargetted and high-throughput 63
screens. A typical narrative is that Scar/WAVE is phosphorylated in response 64
4
to external signalling, through kinases such as (particularly) the global signal 65
transducer ERK2. This has been described in cultured fibroblasts (9), MEFs 66
(10), and endothelial cells (11). The phosphorylation is typically found to 67
change the complex from an inactive to an activatable state, so Scar/WAVE 68
phosphorylation directly leads to actin polymerization. Tyrosine kinases, in 69
particular Abl, have been found to be similarly activating (12, 13). These 70
reports are curious, for a number of reasons. First, actin is a strongly acidic 71
protein, so phosphorylation of binding proteins typically weakens their 72
affinity for actin and actin-related proteins. Second, ERK2 has a tightly 73
defined consensus sequence; however, the proposed phosphorylation sites 74
(and confirmed by us below) do not fit this consensus. We have therefore 75
explored the biological functions of Scar/WAVE phosphorylation in detail. 76
A separate set of phosphorylations is present in the C-terminal WCA domain 77
of Scar/WAVE. It is not detectable by, for example, a change in banding 78
pattern on western blots, and difficult to see by mass spectrometry, so is far 79
less widely described. We (14) and others (15) have shown that this is 80
constitutive, and has a role in tuning the sensitivity of the Scar/WAVE 81
complex rather than activating it; both phosphomimetic and 82
unphosphorylatable mutants are active. 83
One key process in Scar/WAVE biology that is particularly poorly understood 84
is autoactivation. It is clear that pseudopods of migrating cells (which are 85
caused by Scar/WAVE) are controlled through positive feedback – new actin 86
polymerization occurs adjacent to recent pseudopods (8), leading to travelling 87
waves at the edges of cells (16), but the mechanism of this regulation is not 88
well understood (17, 18). It does, however, emphasize the importance of 89
understanding the full dynamics of Scar/WAVE – its recruitment and release 90
from pseudopods, and its synthesis and breakdown – rather than focussing 91
exclusively on its activation. 92
5
93
There is no compelling reason to connect Scar/WAVE phosphorylation to the 94
activation step. Phosphorylation could alter the activity of the complex after 95
it is activated, or alter properties like the rate of autoactivation or the stability 96
of Scar/WAVE once recruited (18). Indeed, in the present work, we find that 97
phosphorylation’s primary role is to control the lifetime of Scar/WAVE after it 98
is recruited to the edge of the cells, meaning that it correlates with pseudopod 99
size rather than the probability of pseudopod generation. We show that 100
phosphorylation occurs after the complex is activated, and at a number of 101
separate, dissimilar sites throughout Scar/WAVE’s polyproline domain. Thus 102
its role appears to be centred around biasing pseudopod behaviour, as 103
required by pseudopod-based models of cell migration, rather than in 104
initiating new pseudopods or actin polymerization. 105
106
6
Results 107
Detailed analysis of Scar/WAVE phosphorylation in vivo 108
A number of authors have observed Scar/WAVE phosphorylation (10, 11, 14, 109
19, 20). Overall, it is clear that there is constitutive phosphorylation in inactive 110
Scar/WAVE, which we have shown is localised in the extreme C-terminus 111
(14). There is also additional phosphorylation seen in (for example) growth 112
factor stimulated cells, which a number of authors have placed downstream 113
of ERK2 (10, 11, 19), implying it is part of the process by which signalling 114
causes pseudopods to form. To analyse the phosphorylation in detail, we 115
devised an optimised electrophoresis regime. Examined by western blot on 116
normal SDS-PAGE gels, Dictyostelium Scar runs as a single band with a 117
diffuse upper edge; quantitative analysis shows this band forms a single, 118
diffuse peak (Figure 1A). To improve separation of different phosphoforms, 119
we optimised PAGE systems. The best separation occurred on 10% 120
acrylamide gels containing low bis-acrylamide concentrations of 0.06% (low-121
bis gels) compared to the usual 0.3%. In such gels, Scar from migrating 122
Dictyostelium resolves into at least six distinct bands (Figure 1B) with clear 123
separation on an intensity plot. Lambda phosphatase resolved the multiple 124
bands to a single one, shifting the multiple intensity peaks (solid line) to 125
single peak (dotted line) (Figure 1C), confirming that the multiple bands are 126
due to phosphorylation, and Scar is typically hyperphosphorylated to varying 127
degrees. 128
Similarly, WAVE2 from two mammalian lines (human MDA-MB-231 and 129
mouse NIH3T3) appears on Western blots as a broad band with a single 130
intensity peak (Figure 1D). As with Dictyostelium Scar, low-bis gels reveal 131
multiple phosphorylated WAVE2 bands (Figure 1E). Intensity plots show 132
7
multiple well-resolved peaks that are shifted down after phosphatase 133
treatment confirming the bands represent WAVE2 phosphoforms. 134
Identification of phosphorylation sites in vivo 135
Previous reports identified Scar/WAVE phosphosites using protein expressed 136
in vitro (11), or overexpressed fusion proteins (9). Because Scar/WAVE’s 137
biological roles occur exclusively as part of the five-membered Scar/WAVE 138
complex, and overexpressed proteins are unevenly and incompletely 139
incorporated into the complex, we analysed the phosphorylation of native 140
proteins. We purified the Dictyostelium Scar complex using nap1- cells stably 141
transfected with single copies of GFP-Nap1 (14), yielding normal expression 142
levels, and pulled down the complex using GFP-TRAP. Under these 143
conditions Scar is only purified if incorporated in a properly assembled 144
Scar/WAVE complex. Gel-purified Scar bands (Supp. Figure 1A &B) were 145
analysed by LC-MS/MS to identify phosphopeptides. Scar/WAVE is 146
composed of N-terminal SH, central B and polyproline, and C-terminal V, C 147
and A domains (21). We identified three phosphorylated tyrosines (Y88, 129, 148
210) in the SH domain, three phosphorylated serines (S287, 290, 301) in the 149
polyproline and three (S384, 388, 389) in the V domain (Figure 1F). These are 150
additional to the constitutive phosphorylations found by Ura et al. from 430-151
440 in the A domain. The number of SCAR forms discernible as bands in 152
western blots from migrating cells is therefore not a surprise – at least 16 sites 153
are phosphorylated in vivo in normal cells. 154
Conspicuously, none of the phosphorylated serines made up MAP kinase 155
consensus sites (Supplementary Table 1). This was surprising, as a number of 156
papers report that Scar/WAVE bands (phosphorylation) are caused by ERK2, 157
which has a strong preference for prolines 2aa N-terminal to and 1aa C-158
terminal to the target site (22). The sites in human WAVE2 do not fit the MAP 159
8
kinase consensus (Supplementary Table 1), with the exception of T346, which 160
is relatively sparsely phosphorylated (https://phosphosite.org). 161
162
To confirm that we had identified the full range of phosphorylations, we 163
tested which amino acids had to be mutated to block extra bands in westerns. 164
We expressed un-phosphorylatable (Y-F; S-A) and phosphomimetic (S-D) 165
mutants in scar- cells. Removing the five phosphorylated serines in the A 166
domain of Scar (14) by deleting the VCA domain (ScardVCA) did not reduce the 167
number of bands (Figure 1G). Mutation of Y88/129/210F in SHD of Scar did 168
not affect the band shifts (Figure 1H). Mutation of the three serines in the 169
polyproline domain (ScarS3A, Figure 1H) to alanines reduced the number of 170
bands but did not abolish them. Mutating two more nearby serines (giving 171
ScarS5A; S284,290,301,335,339) gave near-complete loss of Scar bands (Figure 172
1I). Additional mutation of S384/388/389A from the V domain, giving ScarS8A 173
(Figure 1J), yielded Scar that was nearly homogeneous, with a single very 174
faint band above the main Scar population. The residual heterogeneity was 175
lost in cells whose Scar had also lost the sites in the A domain (14), giving 176
ScarS13A (Figure 1I, J & K). Similarly, mutation of all 5 polyproline 177
serines(ScarS5D), or all 8 polyproline+V domain serines to aspartate (ScarS8D), 178
caused a substantial mobility shift and two distinct bands on low bis gel 179
(Figure 1K). The mobility shift of ScarS8D was also obvious on a normal-bis gel 180
(Supp. Figure 1C), but only a single band of ScarS8D was resolved. These two 181
forms are apparently not caused by phosphorylation; strong phosphatase 182
treatment, which also caused significant proteolysis, caused only partial loss 183
of the upper band (Figure 1L); the difference may be due to a small modifier 184
like ubiquitin, though we were unable to find ubiquitin itself by western blot 185
in any variant (ScarWT, ScarS8A or ScarS8D; Supp. Figure 2). 186
9
The mammalian Scar/WAVE2 has been also shown to be phosphorylated at 187
multiple positions (Figure 1M, Supp. Table 1). Similar to Scar, mutation in 188
WAVE2S8A/T1A (S293/296/298/308/343/351/429/442/T346A) removed all 189
phosphorylated bands (Figure 1N). This confirmed that the multiple bands in 190
WAVE2 are due to S/T phosphorylations. 191
In conclusion, the multiple bands seen in westerns are mostly caused by 192
differential phosphorylation at sites in the polyproline region, with minimal 193
contributions from the V and A domains. 194
195
Scar/WAVE is phosphorylated after activation 196
The Scar/WAVE complex must interact with the GTP-bound, active form of 197
Rac1 before it can catalyse Arp2/3 complex activation and pseudopod 198
generation (3, 23–28). We would not expect Rac interactions to be important to 199
the phosphorylation if, as described in several publications (3, 23, 26), the 200
phosphorylation is an upstream process regulating Scar/WAVE activation. 201
We therefore determined the effect of Rac1 binding on Scar/WAVE 202
phosphorylation. First we inhibited the Rac1 activity in both Dictyostelium 203
and B16F1 cells by EHT1864, an effective inhibitor of RacGEF activity (29, 30), 204
and assessed the band shifts of Scar/WAVE. EHT1864 treatment resulted in 205
complete loss of Pak-CRIB-mRFPmars2 from cell periphery and depolarized 206
Dictyostelium cells (Figure 2A, Suppl. Video 1), confirming that the inhibitor 207
reduced active Rac levels to almost nothing. Similarly, lamellipods of B16F1 208
cells collapsed after EHT1864 treatment; their F-actin at cell edges also 209
appeared reduced (red, Figure 2B). 210
To assess the effect of Rac1 inhibition on Scar phosphorylation, Dictyostelium 211
cells were incubated with 50µM EHT1864 for 2, 5 and 10 min, lysed, and 212
analysed by low-bis western. Inhibition of Rac1 resulted in a strong reduction 213
10
in Scar phosphorylation (Figure 2C). The intensity profile shows the 214
disappearance of peaks for polyphosphorylated Scar even more clearly 215
(Figure 2D). Quantification of the fraction of Scar in the upper bands shows 216
approximately 50% reduction (Figure 2E). This figure underestimates the total 217
change in phosphate, because the higher bands that are disproportionately 218
lost contain several phosphates. Similarly, inhibition of Rac1 in B16F1 cells by 219
EHT1864 reduced the intensity of the upper bands of WAVE2 within 2 h; the 220
change became even clearer after 6 h of treatment (Figure 2F). Intensity plots 221
of WAVE2 bands show strong reduction in the highest peak of WAVE2 (0h; 222
pink) after EHT1864 treatment (Figure 2G). Quantification of the ratios of the 223
upper (intense) and lowest bands also suggest a substantial loss of WAVE2 224
phosphorylation (Figure 2H, again an underestimate of total phosphate loss). 225
This implies that Scar is only phosphorylated after it has bound to Rac1, and 226
thus been activated. 227
Generalized inhibition of Rac1 activity is strongly likely to cause secondary 228
effects, as many proteins such as kinases (for example PAKs (31)) are 229
important for other aspects of motility and require Rac to be activated. To 230
verify that active Rac1’s effect on Scar phosphorylation is direct, we examined 231
cells in which the Pir121 protein was replaced by Pir121-EGFP, either wild 232
type or the Rac1 non-binding, A-site mutant (3, 5). Pak-CRIB-mRFP-mars2 (8) 233
localizes to the pseudopods and cell periphery in both strains (Figure 2I; 234
Panel I, Suppl. Movie 2), showing that Rac is similarly activated in WT and 235
mutant cells. However, the Scar complex is only activated and localized in 236
cells expressing the functional Pir121EGFP, not the A-site mutant (Figure 2I; 237
Panel II, Suppl. Movie 2). Scar phosphorylation is greatly diminished in the A 238
site mutant (Figure 2J), to a level comparable to that seen in Rac-inhibited 239
cells. This confirms that Scar phosphorylation occurs only after the complex is 240
activated by interaction with Rac1. 241
11
242
To affirm the sequence of activation and phosphorylation, we tested the effect 243
of latrunculin, an inhibitor of actin polymerization (32, 33), that causes 244
exaggerated membrane recruitment of the Scar/WAVE complex in 245
mammalian cells (34) and Dictyostelium. In Dictyostelium, latrunculin causes 246
concentrated and persistent recruitment of Scar complex to the membrane 247
(Figure 2K, Suppl. Video 3). Upon latrunculin treatment Scar 248
phosphorylation rapidly increases; like the Scar recruitment, the 249
phosphorylation reaches a far higher level than is normally seen (Figure 2L). 250
Thus Scar phosphorylation again correlates with its recruitment and 251
activation, rather than upstream processes. 252
It is interesting to note that there is currently no biochemical assay for 253
Scar/WAVE activity. The only measure of activation is recruitment to the 254
extreme leading edge, and downstream consequences such as actin 255
polymerization. This phosphorylation assay could therefore be a potentially 256
useful indirect measure of recent Scar/WAVE activation in vivo. 257
Chemotactic signalling does not control Scar phosphorylation 258
Earlier reports described the MAP kinase ERK2, induced downstream of 259
chemotactic and growth factor signalling, as the principal driver of 260
Scar/WAVE phosphorylation (9, 11, 19). This is an important issue, because it 261
provides a clear connection between chemoattractant signalling and 262
pseudopod initiation and growth that is otherwise lacking. As described 263
earlier, we questioned this result, because MAP kinase sites are usually 264
restricted to a narrow consensus, whereas the sites we identified were 265
diverse. We therefore examined the effect of extracellular signals and MAP 266
kinases on Scar/WAVE phosphorylation. Growing Dictyostelium cells use 267
folate as a chemoattractant. During multicellular development they 268
12
downregulate folate receptors and express receptors to extracellular cAMP. 269
Treatment of growing cells with folate, or developed cells with cAMP, causes 270
increases in F-actin levels that are thought to be important for chemotaxis (35, 271
36). To provide a consistent narrative, we examined Scar phosphorylation 272
after both treatments. Despite massive activation of erkB protein, no change in 273
the Scar phosphorylation was seen in either folate- or cAMP-treated cells 274
(Figure 3 A & D), either visually from the gel or when expressed 275
quantitatively as the phosphorylated fraction (Figure 3B, C, E & F). 276
Knockouts of the Dictyostelium Gb protein show a complete loss of G-protein 277
signalling (37) and reduced ErkB activity (38); since there are also no tyrosine 278
kinase receptors in Dictyostelium these cells are complete chemotactic nulls. 279
However, Gb- mutant cells have normal levels of Scar phosphorylation 280
(Figure 3G &H), as well as apparently normal pseudopods (39). This confirms 281
that chemotactic signalling is not required for Scar phosphorylation. 282
We determined the importance of MAP kinase signalling for Scar/WAVE 283
phosphorylation in both Dictyostelium and mammalian cells. First we 284
examined the band shifts for Scar in western blots from erkA-, erkB- and erkAB- 285
nulls of Dictyostelium (40). The band shifts on the western blot and quantitated 286
peaks were not discernibly different from the WT parent (Figure 3I&J). Since 287
ErkA and ErkB are the only MAP kinases in Dictyostelium, MAP kinase 288
signalling is not important for Scar/WAVE phosphorylation. To confirm this 289
result is generally true, we examined the WAVE2 phosphorylation band shifts 290
in mammalian NIH3T3 and B16F1 cells after inhibition of Erk2 activity. 291
Treatment with U0126, an efficient inhibitor of MEK (41) immediately 292
abolished Erk2 phosphorylation (and thus activity) in both NIH3T3 and 293
B16F1 cells, but did not change the phosphorylation state of WAVE2 even 294
after prolonged incubation (Figure 3 K, L, M&N). These results confirm that 295
13
the Scar/WAVE phosphorylations we observe are neither executed by Erk2 296
nor do they require Erk2 function. 297
Physical adhesion promotes Scar/WAVE phosphorylation 298
More generally, our data suggest Scar/WAVE phosphorylation is not 299
primarily regulated by extracellular signalling. Another strong candidate 300
regulator of pseudopod and lamellipod is cell:substratum adhesion. In the 301
absence of adhesion, cell protrusions collapse, and the Scar/WAVE complex is 302
strongly localised to the adherent edges of protrusions. We exploited the 303
ability of Dictyostelium to grow either in suspension or adhesion. First, we 304
compared the differences in Scar phosphorylation in cells grown in 305
suspension and adhesion in both vegetative and starved cells. The more 306
phosphorylated bands (* in Fig. 4A) are more intense in cells grown adherent 307
to Petri plates, when compared with suspension-grown cells. Similarly, 308
intensity plots of Scar from cells adherent to a surface clearly show more 309
peaks and greater intensities of more-phosphorylated peaks compared with 310
cells grown in suspension (Figure 4B). This observation suggested that 311
adhesion is a strong stimulant of Scar phosphorylation. To investigate this in 312
more detail, we allowed suspensions of cells to adhere to a plastic surface, 313
and observed a clear and highly significant (p= 0.0089; Kruskal Wallis test) 314
upregulation of Scar phosphorylation (Figure 4 C&D). The proportion of 315
intensity in phosphorylated bands increases from 25±3.98% to 42.5±1.99% 316
(Mean±SEM; Figure 4D). Conversely, when we examined adherent cells that 317
had been de-adhered by a stream of buffer from a pipette and kept in 318
suspension, we saw an obvious reduction in phosphorylated bands (Figure 319
4E). The proportion of intensity in phosphorylated bands decreased from 320
40±3.59 to 20±3.08 (mean, SEM; Figure 4F). Thus cell:substrate adhesion 321
causes activation of Scar phosphorylation. 322
14
We examined whether cell:substrate adhesion coupled to Scar activation via 323
adhesion-linked signalling, or through a physical process such as 324
deformation. This was enabled by talin A/B double knockout Dictyostelium 325
cells. Talin is a key couple between adhesion molecules like integrins and the 326
actin cytoskeleton (42, 43) and talA/B- mutant cells completely fail to adhere 327
when allowed to fall onto a substratum (44). Consequently they remain 328
extremely spherical and pseudopod-free when cultured, even in dishes where 329
normal cells adhere. However, we were able to force talA/B- cells into 330
adhesion to plastic, glass or 1 % agar surfaces by carefully aspirating the 331
liquid from the dishes so cells were compressed by capillary forces. Scar 332
phosphorylation increases sharply in cells that are making talin-independent 333
contacts with the substratum and the comparison of Scar phosphorylation 334
between suspension and attached talA/B- cells are identical to WT cells (Figure 335
4G). Intensity plots show no difference in phosphorylation between adherent 336
WT cells and talA/B- cells that have been forced to adhere (Figure 4 H, I & J). 337
This result implies that Scar/WAVE phosphorylation is mediated by the 338
physical process of attachment – through mechanochemical processes, for 339
example - rather than through canonical integrin adhesion signalling. 340
To determine if Scar/WAVE2 phosphorylation in mammalian cells is also 341
adhesion dependent, we compared the band shifts in cells that grow in 342
suspension (Jurkat and JVM3) and adhered to a surface (NIH3T3, the 343
pancreatic cancer lines PDACA & B, Cos7, MDA-MB-231 and B16F1). The 344
Jurkat and JVM3 cells have less intense upper bands, meaning highly 345
phosphorylated Scar/WAVE2 is less abundant. Both also have an additional 346
lower band (representing less-phosphorylated Scar/WAVE2) not seen in any 347
of the cells that grow in adhesion (Figure 4K). The intensity ratios of the 348
upper intense and lowest of the bands present in all cells showed a marked 349
increase in adherent cells (Black bars in Figure 4L) compared with suspension 350
15
cells (White bars in Figure 4L). We also modulated adhesion by varying cell 351
density – cells grown at high confluency have a lower adhesion area due to 352
space constraints imposed by neighbouring cells. The amount of Scar/WAVE2 353
phosphorylation decreases with confluency in both NIH3T3 and B16F1 cells 354
(Figure 4M, N & O). 355
Unphosphorylated and Phosphomimetic Scar Mutants are Fully 356
Functional 357
In essentially every lamellipod or pseudopod of every normal cell, actin and 358
Arp2/3 complex recruitment are mediated by the activation of the Scar/WAVE 359
complex (2). Phosphorylation of the polyproline domain has been described 360
as important for Scar/WAVE complex activation and formation of such 361
protrusions (9–11, 14, 19, 20). To determine the importance of polyproline 362
domain phosphorylation to cell migration in vivo, we constructed 363
unphosphorylatable (ScarS8A) and phospho-mimetic (ScarS8D) mutants, with 364
serines mutated to alanines and aspartates, respectively (Figure 1F & J) and 365
expressed them in scar- cells. Both mutants were expressed at wild type levels 366
(Figure 1J). 367
Cells lacking Scar are inefficient in migration, as they have small pseudopods 368
and migrate mainly by blebbing, and are less directional ((8)Figure 5A; Panel 369
I, Suppl. movie 4). In contrast, scar- cells rescued with ScarWT, ScarS8A and 370
ScarS8D formed longer pseudopods that moved faster and split frequently 371
(Figure 5A; Suppl. Video 4). Pseudopod formation was similar in cells rescued 372
with WT and either Scar mutant. The migration speed of knockout cells was 373
completely rescued by expression of wild type or either phosphomutant Scar 374
(Figure 5B), with phosphorylation changes causing a slight increase in cell 375
speed. The tortuosity of the cell perimeters, which describes the amount of 376
recent protrusion and shape change, was low in scar- cells but rescued by 377
16
ScarS8A and ScarS8D just as well as by ScarWT (Figure 5C). Thus, 378
phosphorylation is clearly neither a precondition for activity, nor a direct 379
cause of SCAR inactivation. 380
Unphosphorylatable Scar is Recruited in Larger and Longer-lived 381
Patches 382
Activated Scar complex localizes to the fronts of cells and causes actin 383
protrusions there (8). We have found that GFP-tagging Scar itself greatly 384
alters both Scar and pseudopod dynamics. Therefore to examine the effects of 385
Scar phosphorylation on the dynamics of Scar recruitment, we expressed Scar 386
mutants in Scar- cells in which Nap1 has been replaced by a single copy of 387
GFP-Nap1 (i.e. scar-/Nap1GFP cells; Ura et al., 2012). ScarWT, ScarS8A and ScarS8D 388
all localized efficiently at the pseudopod periphery, but there were marked 389
differences in the dynamics of the association (Figure 5D; Suppl. Video 5). As 390
usual (14) ScarWT was recruited in bursts of fairly short duration (8.2 s ± 5.9s, 391
mean±SD). The unphosphorylated Scar (ScarS8A) remained recruited for a 392
substantially longer duration (13.7s ± 11.63s, mean±SD). As well as being 393
longer-lived (Figure 5E), ScarS8A patches were also larger (Figure 5F). This led 394
to changes in patch frequency – the larger, longer-lived ScarS8A patches were 395
made at a lower frequency than ScarWT (Figure 5G). Conversely, localization 396
of ScarS8D was even briefer (6.37 ±1.94s, mean±SD) and patches were made at 397
an even higher frequency than ScarWT, though they were about the same size. 398
Based on these observations, we hypothesized that Scar/WAVE 399
phosphorylation is not required for its activation, but rather modulates cell 400
migration by controlling the dynamics of Scar/WAVE patches. To confirm 401
this, we constructed nearly completely un-phosphorylated and 402
phosphomimetic Scar by additionally mutating the 5 serine residues of the 403
acidic domain (Ura et al., 2012). Scar complex and F- actin were both 404
17
beautifully accumulated in the pseudopods of cells expressing ScarS13D and 405
ScarS13D (Suppl. Video 6). 406
To further confirm this observation in B16 F1 mouse melanoma cells, we 407
employed WAVE1/2 knock out cells and examined the effect of unphospho- 408
mutations in the proline-rich domain (WAVE2S8A/T1A; Figure 1N). WAVE1/2 409
KO cells transfected with empty vector did not form lamellipodia (Figure 5H). 410
However, expressing both WAVE2WT and WAVE2S8A/T1A in knockout cells 411
rescued lamellipodia formation. These results firmly indicate that Scar 412
phosphorylation is neither required for its activation, nor does it cause direct 413
inactivation. 414
Scar Phosphomutants Rescue scar-/wasp- Cells 415
In Dictyostelium cells lacking Scar, WASP can be repurposed to drive 416
pseudopod formation and Arp2/3 complex activation (8, 45), which 417
complicates the phenotypes of scar mutants. Scar-/wasp- cells cannot move 418
and cannot grow; we have therefore developed a tetracycline-inducible 419
double knockout (scartet/wasp-) cell line to test Scar function without WASP 420
complementation (45). We exploited this line to test unphosphorylatable Scar 421
mutants’ ability to support growth and pseudopod formation. Scartet/wasp- 422
cells were transfected with ScarS8A and ScarS8D then deprived of doxycycline 48 423
h prior to experiments to remove native Scar. As usual (46) both mutants 424
express at wild type levels (Figure 6A). Importantly, expression of both 425
ScarS8A and ScarS8D is dominant; expression of the native Scar is suppressed 426
even in the presence of doxycycline (Figure 6A). 427
Similar to ScarWT, both ScarS8A and ScarS8D rescued the growth of Scartet/wasp- 428
cells (Figure 6B). To study the migratory phenotype, cells were examined 429
migrating under agarose up a folate gradient. As previously observed (45), 430
repressed Scartet/wasp- cells made long, stable filopods, but could not form 431
18
pseudopods or migrate. However, cells expressing either ScarS8A or ScarS8D 432
showed clear localisation of Scar complex to protrusions and F-actin 433
polymerisation (Figure 6 C&D; Suppl. Video 7), and robust pseudopod 434
formation (Figure 6E; Suppl. Video 8). Both mutants rescued cell migration 435
effectively (scar-/wasp- cells do not migrate under the agar, so their speed 436
cannot be measured in this assay, but it is essentially zero). ScarS8A supported 437
a substantially higher speed (8.35 ± 3.2µm/min, mean±SD) than either WT 438
(3.71 ± 2µm/min, mean±SD) or ScarS8D (5.2 ± 2.1 µm/min, mean±SD; Figure 6F). 439
These results firmly indicate that unphosphorylatable and phosphomimetic 440
mutants are fully functional for growth and cell migration in cells without 441
requiring cooperation from WASP. 442
Phosphorylation as a Degradation Signal 443
Scar/WAVE complex dynamics at the pseudopod periphery are poorly 444
understood. In particular, the regulation of Scar/WAVE removal – which 445
must be rapid, to allow the dynamic behaviour observed in moving cells - is 446
mysterious (24). Our results suggest that phosphorylation is among the 447
mechanisms that control the rapid changes in Scar complex localization (Figs 448
2K, 5 and 6). One way this could be achieved is by promoting breakdown of 449
the complex. This seems particularly appropriate as we previously found that 450
mutations in the acidic domain rendered Scar unstable (14) as well as 451
excessively active. Furthermore, re-examining our earlier data (Figures 4D 452
and 4F, for example) we noticed a surprising change – under conditions 453
where Scar phosphorylation decreases, such as deadhesion, there is an acute 454
increase in the total amount of protein (Figure 7A). Similarly a drop in Scar 455
phosphorylation causes protein levels to increase acutely (Figure 7B). We 456
therefore compared the stability of ScarWT, ScarS8A and ScarS8D during 457
starvation, with and without cycloheximide to inhibit protein synthesis. 458
19
Starvation makes cells move more rapidly (47) and increases Scar levels 459
(Figure 7 C&D). In the absence of cycloheximide, ScarWT & ScarS8A levels 460
increased as usual during starvation (Figure 7 C&D). ScarS8D levels, however, 461
decreased (Figure 7C&D). When cells were starved with cycloheximide, 462
amounts of ScarWT, ScarS8A and ScarS8D all decreased, but the loss was 463
substantially stronger in ScarS8D (Figure 7C & E). This result shows that 464
phosphomimetic Scar is less stable, implying that the reason ScarS8D makes 465
smaller and shorter-lived pseudopods is that it is degraded more rapidly. 466
20
Discussion 467
We have used an improved assay for phosphorylation of Scar/WAVE’s 468
polyproline domain to open a window on pseudopod dynamics. This has 469
revealed a number of surprises. Phosphorylation occurs on multiple sites 470
after Scar/WAVE is activated by proteins such as activated Rac, and reveals 471
an adhesion-dependent, signalling-independent pathway. It is clear that 472
these phosphorylations are not important for Scar/WAVE to be activated, and 473
it seems likely that they do not inhibit its activation, either. Rather, 474
phosphorylation appears to be a relatively subtle tool by which cells can 475
manipulate the dynamics (both recruitment and loss) of Scar/WAVE patches 476
and thus pseudopods. Pseudopod lifetime intersects with a wide range of cell 477
biology – it underlies the mechanism of chemotactic steering (48, 49), and 478
regulates cell polarity, persistence (50) and shape change (51) – and mutants 479
with uncontrolled pseudopod lifetime have serious motility defects (4). Thus 480
although the function of Scar/WAVE phosphorylation is different from what 481
has been believed, it is fundamentally physiologically important. 482
483
Our finding that signalling does not greatly alter the rate of Scar/WAVE 484
phosphorylation, and thus activation, does not imply that signalling is not 485
important to pseudopod formation and evolution. Chemotaxis – cells 486
migrating up gradients of soluble signalling molecules – clearly happens, and 487
is clearly mediated by pseudopods. However, we (48) and others (49) have 488
found that normal chemotaxis rarely involves initiation of new pseudopods – 489
rather, pseudopod generation is mostly random, but localised receptor 490
activation modulate the positions where pseudopods evolve, their lifetime 491
and stability after they are formed, and sometimes the pattern in which new 492
split pseudopods are directed. None of these measured processes require 493
21
extracellular signals causing direct activation of pseudopod catalysts; rather, 494
more subtle changes like the precise subcellular localisation and timing of 495
Scar/WAVE activation could be biased by chemoattractant signalling. 496
Equally, chemoattractants could work through pathways other than 497
Scar/WAVE activation, biasing the growth and lifetimes of pseudopods rather 498
than their initiation. This is in full agreement with the pseudopod-centred 499
view of chemotaxis (52), which emphasizes that pseudopods are self-500
organized, rather than the outcome of external signals. 501
502
The kinase sites around the polyproline domain have little in common with 503
one another, and do not match the consensus sequences for most kinases that 504
are active in the actin domain. We hypothesize that the phosphorylation sites 505
are sterically inaccessible until the Scar/WAVE complex is activated. Once the 506
complex is active it adopts an open conformation, whereupon any and all of a 507
large range of constitutive kinases such as CK2 and GSK3 could act on it. In 508
this case the phosphorylation would be a passive reflection of the fact that the 509
complex had been activated, rather than a driver or even modulator of the 510
initial activation process. This explanation is comfortably consistent with the 511
wide range of data we have obtained, though it paints a different picture of 512
Scar/WAVE activation from what we had expected. 513
514
This work showcases our ability to use scar-/wasp- cells to test the functioning 515
of mutated Scar proteins, and the phenotypic changes caused by alterations in 516
the sequence. Previously, results have been complicated because WASP 517
changes its organization and cell biological role when the Scar complex is not 518
present (8). We have therefore been working against an ill-defined 519
background of partial rescue. In this work we show that either 520
phosphomimetic or unphosphorylatable Scar is fully able to rescue the 521
22
mutant phenotypes, because scar-/wasp- Dictyostelium cells cannot grow, do 522
not make pseudopods, and barely move. Similarly, the use of 523
WAVE1/WAVE2 knockout B16-F1 cells allows detailed study of 524
physiologically normal levels of WAVEs without partial complementation. 525
We therefore look forward to a new focus on physiological function and a 526
more direct connection between the dynamics of the Scar/WAVE complex, the 527
behaviour of pseudopods and lamellipods, and cell migration. 528
529
Our results imply that cell:substrate adhesion is a crucial driver of 530
Scar/WAVE activation. Adhesion has been an undeservedly neglected area of 531
Scar/WAVE biology. It seems intuitively obvious that pseudopod formation 532
should be regulated by adhesion – if pseudopods or lamellipods are 533
generated in a location where adhesion is difficult or impossible, they will be 534
inefficient or ineffective at producing cell migration. Cells should in 535
retrospect therefore be expected to monitor adhesion, and to favour 536
pseudopods that can attach. The results in Figure 4 showing normal 537
Scar/WAVE activation in talin knockouts offer one reason why connections 538
have so far been missed – activation does not require the integrin-linked 539
signalling pathways that dominate adhesion of (for example) mammalian 540
fibroblasts (53). Instead, we hypothesize that the connection is physical, by 541
direct mechanical coupling. This is a well-documented and common 542
mechanism for coupling proteins and cell behaviour (54). Overall, this work 543
exposes a new and conserved mechanism for regulating migration, which will 544
influence cells moving in many physiological contexts. We look forward to 545
seeing new biological examples emerge. 546
547
23
548
549
24
Acknowledgements 550
We thank Prof. Robert R. Kay for erkA and erkB knockout cells, Dr. Rachana 551
Patel for providing NIH3T3 cells, Drs Margaret O’Prey, Heather Spence and 552
Jamie Whitelaw for help in microscopy and mammalian cell culture, and Dr. 553
Simona Buracco for discussions about protein stability. 554
Funding 555
This work was supported by Cancer Research UK core grant number A17196 556
and Multidisciplinary Award A20017 to RHI, by the Deutsche 557
Forschungsgemeinschaft, grant GRK2223/1 to KR, and a grant from the NIH 558
(GM063691) to BG. 559
560
Authors’ contribution 561
RI and SPS conceived and designed experiments. SPS performed experiments. 562
PT assisted SPS during in experiments and provided essential reagents. SL 563
performed MS. SPS wrote the manuscript. RI and LM corrected and improved 564
the manuscript. MS, QT, BG and KR generated B16-F1 mutant cell lines and 565
plasmids 566
Declaration of Interests 567
The authors declare no competing interests 568
569
25
Figure Legends 570
Figure 1: Multiple phosphorylations in Scar/WAVE. 571
A) Western blot of Dictyostelium Scar using normal (0.3%) bis-acrylamide gels. 572
Whole cells were boiled in sample buffer and immediately separated on 10% 573
bis-tris gels, blotted and probed with an anti-Scar antibody, yielding a single 574
diffuse band. Scan of signal intensity shows a single, slightly diffuse peak. 575
Mccc1 is a loading control (55). 576
B) Western blot of Dictyostelium Scar using low (0.06%) bis-acrylamide gels. 577
Whole cells were boiled in sample buffer and immediately separated on 10% 578
bis-tris gels, yielding a multiple discrete bands. Scan of signal intensity shows 579
separate peaks. 580
C) Phosphatase treatment. Whole Dictyostelium cells were lysed in TN/T 581
buffer, then incubated with & without lambda phosphatase, boiled in sample 582
buffer, and analysed using low bis-acrylamide gels. The scan shows multiple 583
peaks resolved into one and shifted downwards after phosphatase treatment. 584
D) Western blot of mammalian WAVE2 in mammalian cells using normal 585
(0.3%) bis-acrylamide gels. Whole MDA-MB-231 and NIH 3T3 cells were 586
boiled in sample buffer, separated on 10% bis-tris gels, blotted and probed 587
with an anti-WAVE2 antibody, yielding a single band. Scan of signal intensity 588
shows a single peak. 589
E) Western blot of mammalian WAVE2 in mammalian cells using low (0.06%) 590
bis-acrylamide gels. Whole MDA-MB-231 and NIH 3T3 cells were boiled in 591
sample buffer and separated on 10% bis-tris gels, yielding at least 5 bands. 592
Incubation with lambda phosphatase resolves the bands into lower bands. 593
Scans of signal intensity show clear multiple peaks resolving incompletely 594
towards one lower peak. All experiments were repeated thrice with 595
comparable results. 596
26
F) Schematic of Dictyostelium Scar architecture and identified phosphorylated 597
residues. 598
G-K) Deletions of different amino acids. Lysates from Scar- cells expressing 599
ScarWT and Scar lacking the VCA domain (G), with Tyr 88, 129 & 201 600
substituted with F (H), Ser 287/290/301 substituted with Ala (H), Ser 601
287/290/301/335/339 substituted with Ala or Asp (I), Ser 602
287/290/301/335/339/384/388/389 substituted with Ala or Asp (J), and Ser 603
287/290/301/335/339/384/388/389/430/431/432/437/440 substituted with Ala or 604
Asp (K), were boiled in sample buffer and analyzed on low (0.06%) bis-605
acrylamide 10% bis-tris gels. Extra bands in Scar are completely lost after 606
substitution of 13 serines. 607
L) Phosphatase treatment of ScarWT, ScarS8A and ScarS8D. Lambda phosphatase 608
treated cell lysates were analysed using low (0.06%) bis-acrylamide 10% bis-609
tris gels. Unlike ScarWT, ScarS8D shows little removal of additional bands after 610
phosphatase treatment. 611
M) Schematic of human WAVE2 architecture and phosphorylated Ser/Thr 612
residues identified in published and high-throughput screens. 613
N) Loss of extra WAVE2 bands in B16F1 mouse cells after substitution of 614
polyproline serines. WAVE1/2KO cells were transfected with intact WAVE2 615
or WAVE2 with Ser 293/296/298/308/343/351/429/442 and Thr 346 substituted 616
with alanine, then lysed, boiled in sample buffer, separated using low (0.06%) 617
bis-acrylamide 10% bis-tris gels, blotted and probed with an anti-WAVE2 618
antibody. 619
620
621
622
Figure 2: Rac1 binding and Scar/ WAVE phosphorylation. 623
27
A) Effect of Rac1 inhibition by EHT 1864 on Pak-CRIB-mRFPmars2 624
localization in Dictyostelium. Dictyostelium cells expressing Rac1 marker, Pak-625
CRIB-mRFPmars2 were imaged by AiryScan confocal microscopy. Pak-CRIB-626
mRFPmars2 (Red) rapidly delocalises after addition of EHT1864 (EHT1864, 50 627
µM) treatment (Scale bar= 10 µm). 628
B) Effect of Rac1 inhibition by EHT 1864 on lamellipod formation in B16 629
melanoma cells. B16F1 cells were seeded on Laminin A coated glass, and 630
treated with EHT1864 (50 µM) for 6h. Cells were fixed and stained with Alexa 631
fluor 568-phalloidin. Untreated cells retained broad lamellipods with intense 632
F- actin at the leading edge, while lamellpods and F-actin are compromised in 633
the EHT1864 treated cells. 634
C-E) Effect of Rac1 inhibition by EHT 1864 on Dictyostelium Scar 635
phosphorylation. Cells were treated with 50 µM EHT1864 for the indicated 636
times, and lysates were analysed for Scar band shifts by western blotting 637
using low (0.06%) bis-acrylamide gels. Disappearance of extra bands after 638
EHT 1864 treatment indicates loss of phosphorylated Scar. Graph in D shows 639
density quantitation of western blots at different times. E shows ratio of 640
upper (more phosphorylated) band aggregate intensity to total Scar (bars 641
show mean±SEM, n=3). 642
F-H) Effect of Rac1 inhibition by EHT 1864 on B16 melanoma WAVE2 643
phosphorylation. Cells were treated with 50 µM EHT1864 for the indicated 644
times, and lysates were analysed for WAVE2 band shifts by Western blotting 645
using low (0.06%) bis-acrylamide gels. Disappearance of extra bands after 646
EHT 1864 treatment indicates loss of phosphorylated WAVE2. Graph in G 647
shows density quantitation of western blots at different times. H shows ratio 648
of upper (more phosphorylated) band aggregate intensity to total WAVE2 649
(bars show mean±SEM, n=3). 650
651
28
I) Requirement for Rac1-binding PIR121 A site for Scar complex localization 652
in Dictyostelium cells. Pir121- cells expressing Pir121-eGFP with and without 653
mutated A site (K193D/R194D) were coexpressed with Pak-CRIB- RFPmars2 654
and imaged by confocal microscopy while migrating under agarose up a 655
folate gradient. Upper panel: unaltered PIR121-eGFP; lower panel: Pir121-656
eGFPA site. Rac1 localizes (Red) to the membrane in both Pir121-EGFPWT and 657
Pir121-EGFPK193D/R194D. Scar complex (green) is recruited to sites of Rac1 658
activation only when Rac can bind. 659
J) Requirement for Rac1-binding PIR121 A site for Dictyostelium Scar 660
phosphorylation. WT, Pir121- and Pir121- cells expressing Pir121 with and 661
without mutated A site (K193D/R194D) were analysed for WAVE2 band shifts 662
by Western blotting using low (0.06%) bis-acrylamide gels. Phosphorylated 663
Scar bands are absent when PIR121’s A site is mutated. 664
K) Localization, distribution and hyperactivation of Scar complex after 665
latrunculin A treatment. Dictyostelium cells expressing eGFP-NAP1 were 666
imaged using an AiryScan confocal microscope. 5µM latA was added at t=0. 667
Scale bar = 10 µm. Scar complex activation at the cell membrane increases 668
after addition of Latrunculin. 669
L) Increased Scar phosphorylation after latrunculin A treatment. Dictyostelium 670
cells were treated with latA for the indicated times, then Scar band shifts were 671
analyzed by Western blotting using low (0.06%) bis-acrylamide gels. 672
Phosphorylated bands become markedly more abundant after latA treatment. 673
All experiments were repeated three times. 674
675
Figure 3: Signalling and Erk independent phosphorylation of Scar/WAVE. 676
A-C) Folate stimulation and Scar phosphorylation in Dictyostelium. Washed 677
growing cells were treated with 50µM folate for the indicated times, then Scar 678
band shifts were analyzed by Western blotting using low (0.06%) bis-679
29
acrylamide gels. Folate does not enhance Scar phosphorylation (B; bands 680
show mean ±SD, n=3) despite substantial signalling response shown by erkB 681
phosphorylation (C). 682
D-F) cAMP stimulation and Scar phosphorylation in Dictyostelium. 683
Aggregation-stage cells were treated with 50µM folate for the indicated times, 684
then Scar band shifts were analyzed by Western blotting using low (0.06%) 685
bis-acrylamide gels. cAMP does not enhance Scar phosphorylation (E) despite 686
very substantial signalling response shown by erkB phosphorylation (bars 687
show mean±SD; n=3). 688
G,H) Scar phosphorylation in WT and Gβ- cells. WT and Gβ- cell band shifts 689
were analyzed by Western blotting using low (0.06%) bis-acrylamide gels. 690
Both show similar band shifts of Scar, as reflected in the quantitation of the 691
upper band (H; bars show mean±SD; n=3) 692
I) Scar phosphorylation in MAP kinase (ERK) mutants. Scar-, WT, erkA-, erkB- 693
and erkAB- cells were analyzed by Western blotting using low (0.06%) bis-694
acrylamide gels. Single mutants show similar phosphorylation; double 695
mutant shows more, as shown by densitometry of the lanes (J). 696
K-N) WAVE2 phosphorylation in B16 (K,L) and NIH3T3 (M,N) cells after 697
inhibition of Erk activation. Cells were treated with U0126 for the indicated 698
times, then analyzed by Western blotting using low (0.06%) bis-acrylamide 699
gels. Erk and phospho-Erk levels were analyzed using normal gels and 700
antibodies against pan-Erk and phospho-Erk. Quantitation of the 701
phosphorylated bands shows no change despite essentially complete 702
inhibition of Erk (bars show mean±SD, n=3). 703
704
Figure 4: Physical adhesion enhances Scar/ WAVE phosphorylation. 705
A,B) Wild type Dictyostelium cells were grown in shaken suspension or 706
adhering to Petri dishes, then analyzed by Western blotting using low (0.06%) 707
30
bis-acrylamide gels. Adherent cells have more and intense bands of 708
Scar/WAVE. Densitometry of lane intensities (mean±SEM, n=3) shows a large 709
difference in the intensities of phosphorylated Scar bands. 710
C,D) Acute effect of adhesion on Scar/WAVE phosphorylation. Cells growing 711
in suspension were allowed to adhere in Petri dishes and lysed at indicated 712
time points. Cell lysates were analyzed by Western blotting using low (0.06%) 713
bis-acrylamide gels. The number and intensity of bands (*) increases with 714
time of adhesion. Quantitation of lane intensities (mean±SEM, n=3) confirms a 715
progressive increase in the intensities of phosphorylated Scar bands. 716
E,F) Acute effect of de-adhesion on Scar/WAVE phosphorylation. Cells grown 717
in Petri dishes were dislodged by a jet of medium from a P1000 pipette and 718
maintained in suspension (2x106 cells/ml) by shaking at 120 rpm. Cell lysates 719
were analyzed by Western blotting using low (0.06%) bis-acrylamide gels. 720
Quantitation of lane intensities (mean±SEM, n=3) confirms a progressive loss 721
in intensity of phosphorylated Scar bands. 722
G-J) Relative requirement of Talin and adhesion. WT and talA-/ talB- cells were 723
allowed to make physical contact with plastic Petri dishes, and where 724
appropriate forced to adhere by withdrawing the liquid with 3MM paper. 725
Cell lysates were analyzed by Western blotting using low (0.06%) bis-726
acrylamide gels. Densitometry of lane intensities shows a large difference 727
between WT and mutant cells on Petri dishes (I), but not when adhesion was 728
forced (J). 729
K,L) WAVE2 phosphorylation in adherent and nonadherent mammalian 730
cells. Lysates from cells that grow in suspension (Jurkat and JVM3) and 731
adhesion (NIH3T3, PDAC-A, PDAC-B, Cos7, MDA-MB231 and B16F1) lysates 732
were analysed for WAVE2 band shifts by Western blotting using low (0.06%) 733
bis-acrylamide gels. Suspension-growing cells have additional lower bands 734
(arrow) and less intense phosphorylated (*) band compared to adherent cells. 735
31
Graph (L) shows the ratio of upper intense band (*) and lowest band of 736
various mammalian cells. 737
M-O) Effect of culture density on WAVE2 phosphorylation. NIH3T3 and 738
B16F1 cells were grown at indicated cell densities and lysates were analysed 739
for WAVE2 band shifts by Western blotting using low (0.06%) bis-acrylamide 740
gels. WAVE2 phosphorylation decreases as the cell density increases. * 741
indicates phospho-WAVE2 bands. Graphs show ratio of upper intense band 742
(*) and lowest band. All graphs are representative of three independent 743
experiments. 744
745
Figure 5: Scar/WAVE phosphorylation mutants are functional. 746
A-C) Rescue of pseudopod formation by mutated Scar. Scar- cells were 747
transfected with ScarWT, ScarS8A and ScarS8D, and allowed to migrate under 748
agarose up a folate gradient while being observed by DIC microscopy. All 749
rescued cells formed actin pseudopods; those formed by ScarS8A expressing 750
cells were broad and elongated. Panel B shows quantitation of speeds and C 751
the tortuosity (total distance migrated/net distance migrated) of paths in B 752
(mean ± SD; n= 140Scar- 170 WT, 413S8A, 437S8D over 3 independent experiments; 753
****p≤0.0001, * p≤0.05, one-way ANOVA, Dunn’s multiple comparison test) 754
D) Subcellular localization of Scar complex. ScarWT, ScarS8A and ScarS8D were 755
expressed in Scar-/Nap1-/eGFP-NAP1 cells, allowed to migrate under agarose 756
up a folate gradient, and examined by AiryScan confocal microscopy. All 757
show efficient Scar/WAVE complex (green) localization to the pseudopods. 758
E,F,G) Lifetimes, size and generation rates of mutant Scar patches. Scar-759
/Nap1-/eGFP-NAP1 cells expressing ScarWT, ScarS8A and ScarS8D were allowed 760
to migrate up folate gradients under agarose, EGFP localization was observed 761
by AiryScan confocal microscopy (1f/3s). Patch lifetime was measured by 762
counting the number of frames a patch showed continuous presence of 763
32
labelled Scar complex, size was measured by hand from single frames, and 764
generation rate calculated from the number of patches lasting at least 2 765
frames. ScarS8A shows increased and ScarS8D diminished lifetime compared 766
with ScarWT . ScarS8A patches are longer-lived than ScarS8D and ScarWT , and 767
ScarS8A generates more and ScarS8D less frequently than ScarWT (mean ± SD; 768
n>25 cells; * p≤0.05, ** p≤0.01, **** p≤0.0001, one-way ANOVA, Dunn’s 769
multiple comparison test). 770
H) Rescue of lamellipodia formation by WAVE2 phosphomutants. B16F1-771
WAVE1-/2- cells expressing WAVE2WT and WAVE2S8A/T1A were allowed to 772
adhere on laminin A coated coverslips, fixed and stained with Alexa 568-773
phalloidin. WAVE2WT and WAVE2S8A/T1A both rescue lamellipodia formation. 774
775
Figure 6: Scar phosphomutants rescue growth and pseudopods in scar-776
/wasp- cells. 777
A) Expression of ScarWT, ScarS8A and ScarS8D in cells lacking endogenous Scar 778
and WASP. scartet/wasp- cells expressing ScarWT, ScarS8A and ScarS8D were 779
grown with or without doxycycline and analysed by Western blotting using 780
low (0.06%) bis-acrylamide gels. MCCC1 was used as a loading control. 781
B) Rescue of Scar/WASP mutant cell growth. scartet/wasp- cells were 782
transfected with ScarWT, ScarS8A and ScarS8D, expression of endogenous Scar 783
was blocked by removing doxycycline at t=0, and cell growth in suspension 784
was measured using a Casy cell counter (Innovatis). Points show mean±SEM, 785
n=3. 786
C) Localisation of Scar complex in scartet/wasp- cells expressing Scar 787
phosphomutants. ScarWT, ScarS8A and ScarS8D were co-expressed with 788
HSPC300-EGFP in Scartet/WASP- cells, and cells migrating under agarose up a 789
folate gradient were imaged by AiryScan confocal microsopy. 790
33
D) Localisation of F-actin in scartet/wasp- cells expressing Scar 791
phosphomutants. ScarWT, ScarS8A and ScarS8D were co-expressed with LifeAct-792
mRFPmars2 in scartet/wasp-, grown with doxycycline to maintain normal Scar, 793
then kept without doxycyline for 48hrs so only the mutant was expressed. In 794
cells without Scar or WASP, LifeAct was observed in thin spiky filament, but 795
the pseudopods of ScarWT, ScarS8A and ScarS8D expressing cells contained 796
normal F-actin levels. 797
E) Rescue of pseudopod formation by mutated Scar. scartet/wasp- cells were 798
transfected with ScarWT, ScarS8A and ScarS8D, and allowed to migrate under 799
agarose up a folate gradient while being observed by DIC microscopy. All 800
rescued cells formed actin pseudopods; those formed by ScarS8A expressing 801
cells were broad and elongated. ScarWT, ScarS8A and ScarS8D expressing cells 802
formed pseudopods. Panel F shows quantitation of speeds observed by 10 X 803
phase contrast microscopy). Almost no unrescued scartet/wasp- cells migrated 804
under the agar so no speed could be measured. (mean ± SD; n>278 cells from 3 805
independent experiments; **** = p≤0.0001, one-way ANOVA, Dunn’s multiple 806
comparison test). 807
808
Figure 7: Phosphorylation destabilizes Scar/WAVE. 809
A) Total Scar levels from adherent cells dislodged from the surface and 810
maintained in suspension. Data measured by densitometry from cells in Fig. 811
4F. Deadhesion, which diminishes phosphorylation, acutely increases SCAR 812
levels. 813
B) Total Scar levels from nonadherent cells acutely allowed to adhere to a 814
surface. Data measured by densitometry from cells in Fig. 4D. Adhesion, 815
which causes a strong increase in phosphorylation, acutely decreases SCAR 816
levels. 817
34
C-E) Reduced stability of phosphomimetic Scar. Scar- cells expressing ScarWT 818
ScarS8A and ScarS8D were starved with and without cycloheximide. Protein 819
samples were analysed for Scar expression at indicated time points using 820
normal 10% bis-tris gels. The graphs show densitometric quantitation of the 821
Scar bands from panel C, with and without cycloheximide (mean±SEM, n=3). 822
823
824
35
Supplementary Figure Legends 825
Supplementary Figure 1: GFP-TRAP pull down of EGFP-NAP1 cells. 826
A) Coomassie brilliant blue stained low-bis acrylamide PAGE gel of GFP-827
TRAP pull down samples. Lane 1,2,3 indicates samples from Scar-828
/EGFPNAP1 and lane 4, 5, 6 indicates Samples from EGFPNap1 cells. The Scar 829
band indicated on the gel was exiced for LC-MS/MS. 830
B) Representative western blot of above gel indicating phosphorylated Scar 831
bands. 832
833
Supplementary Figure 2: Incorporation of Scar mutants in the complex. 834
A) GFP-TRAP pull down was performed using MG132 treated and untreated 835
Dictyostelium Scar-/EGFP-NAP1 cells expressing ScarWT, ScarS8A and ScarS8D. 836
The eluates were analysed on low-bis PAGE western blotting for 837
Pir121/Nap1/Scar and ubiquitin. 838
B) Scar expression in the cell lysates was analysed by low-bis PAGE and 839
western blotting. 840
841
842
Video 1. Effect of EHT1864 on Rac1 and Scar complex localization. 843
Dictyostelium cells expressing PakCRIB-mRFPmars2 were allowed to migrate 844
under agarose up folate gradient and observed by AiryScan confocal 845
microscopy. Filmed at 1 frame/2s, Movie shows 10 frames/sec. EHT1864 was 846
added at frame 7 (after14 secs) in the video. 847
848
Video 2: Scar/WAVE and Rac1 activation in cells with mutant PIR121 A site. 849
Pir121 knock out cells expressing WT Pir121-EGFP and A site Pir121-EGFP 850
were further expressed with PakCRIB-mRFPmars2. Scar complex (green) and 851
36
PakCRIB-mRFPmars2 (Red) localization was visualized in migrating cells 852
under agarose up folate gradient. Filmed at 1 frame/2s, Movie shows 10 853
frames/sec. 854
855
Video 3: Effect of Latrunculin treatment on Scar complex localization. 856
eGFP-NAP1 cells were seeded on LabTekII rcover glass chambers and imaged 857
by AiryScan imaging. 5µm LatrunculinA was added to the cells undergoing 858
imaging. Filmed at 1 frame/15s, Movie shows 10 frames/sec. Latrunculin was 859
added at the frame 2 (after 15 sec). 860
861
Video 4: Pseudopod formation in Scar phosphomutants. 862
Scar- cells expressing ScarWT, ScarS8A and ScarS8D were allowed to migrate 863
under agarose up a folate gradient and observed by DIC. Filmed at 1 864
frame/2s, Movie shows 10 frames/sec. 865
866
Video 5: Scar complex localization in Scar phosphomutants. 867
Scar-/EGFP-Nap1 cells expressing ScarWT, ScarS8A and ScarS8D were allowed to 868
migrate under agarose up folate gradient and Scar complex activation in 869
pseudopods were observed by airyScan confocal microscopy. Filmed at 1 870
frame/2s, Movie shows 10 frames/sec. 871
872
Video 6: Scar complex activation in total Scar phosphomutants. 873
Scar-/EGFP-Nap1 cells expressing ScarS13A and ScarS13D were allowed to 874
migrate under agarose up folate gradient and Scar complex activation in 875
pseudopods were observed by airyScan confocal microscopy. Filmed at 1 876
frame/2s, Movie shows 10 frames/sec. 877
878
37
Video 7: Scar complex activation in Scar-/wasp- cells expressing 879
phosphomutant Scar. 880
Scartet/wasp- cells expressing ScarWT, ScarS8A and ScarS8D were allowed to 881
migrate under agarose up folate gradient and Scar complex activation in 882
pseudopods were observed by AiryScan confocal microscopy. Filmed at 1 883
frame/2s, Movie shows 10 frames/sec. 884
885
Video 8: Pseudopod formation in Scar-/wasp- cells expressing 886
phosphomutant Scar. 887
Scartet/wasp- cells expressing ScarWT, ScarS8A and ScarS8D were allowed to 888
migrate under agarose up folate gradient and were observed by differential 889
interference contrast microscopy. Filmed at 1 frame/2s, Movie shows 10 890
frames/sec. 891
892
38
References 1. Krause M, Gautreau A. Steering cell migration: lamellipodium dynamics and the
regulation of directional persistence. Nat Rev Mol Cell Biol. 2014;15: 577–590.
2. Insall RH, Machesky LM. Actin dynamics at the leading edge: from simple machinery to complex networks. Dev Cell. 2009;17: 310–322.
3. Chen B, Chou HT, Brautigam CA, Xing W, Yang S, Henry L, et al. Rac1 GTPase activates the WAVE regulatory complex through two distinct binding sites. Elife. 2017;6:
4. Fort L, Batista JM, Thomason PA, Spence HJ, Whitelaw JA, Tweedy L, et al. Fam49/CYRI interacts with Rac1 and locally suppresses protrusions. Nat Cell Biol. 2018;20: 1159–1171.
5. Schaks M, Singh SP, Kage F, Thomason P, Klünemann T, Steffen A, et al. Distinct Interaction Sites of Rac GTPase with WAVE Regulatory Complex Have Non-redundant Functions in Vivo. Curr Biol. 2018;28: 3674–3684.e6.
6. Machacek M, Hodgson L, Welch C, Elliott H, Pertz O, Nalbant P, et al. Coordination of Rho GTPase activities during cell protrusion. Nature. 2009;461: 99–103.
7. Yang HW, Collins SR, Meyer T. Locally excitable Cdc42 signals steer cells during chemotaxis. Nature cell biology. 2016;18: 191.
8. Veltman DM, King JS, Machesky LM, Insall RH. SCAR knockouts in Dictyostelium: WASP assumes SCAR’s position and upstream regulators in pseudopods. J Cell Biol. 2012;198: 501–508.
9. Danson CM, Pocha SM, Bloomberg GB, Cory GO. Phosphorylation of WAVE2 by MAP kinases regulates persistent cell migration and polarity. J Cell Sci. 2007;120: 4144–4154.
10. Nakanishi O, Suetsugu S, Yamazaki D, Takenawa T. Effect of WAVE2 phosphorylation on activation of the Arp2/3 complex. J Biochem. 2007;141: 319–325.
11. Mendoza MC, Er EE, Zhang W, Ballif BA, Elliott HL, Danuser G, et al. ERK-MAPK Drives Lamellipodia Protrusion by Activating the WAVE2 Regulatory Complex. Mol Cell. 2011;41: 661–671.
12. Stuart JR, Gonzalez FH, Kawai H, Yuan ZM. C-ABL interacts with the wave2 signaling complex to induce membrane ruffling and cell spreading. J Biol Chem. 2006;281: 31290–31297.
39
13. Leng Y, Zhang J, Badour K, Arpaia E, Freeman S, Cheung P, et al. Abelson-interactor-1 promotes WAVE2 membrane translocation and Abelson-mediated tyrosine phosphorylation required for WAVE2 activation. Proc Natl Acad Sci U S A. 2005;102: 1098–1103.
14. Ura S, Pollitt AY, Veltman DM, Morrice NA, Machesky LM, Insall RH. Pseudopod Growth and Evolution during Cell Movement Is Controlled through SCAR/WAVE Dephosphorylation. Curr Biol. 2012;22: 553–561.
15. Pocha SM, Cory GO. WAVE2 is regulated by multiple phosphorylation events within its VCA domain. Cell Motil Cytoskeleton. 2009;66: 36–47.
16. Killich T, Plath PJ, Wei X, Bultmann H, Rensing L, Vicker MG. The locomotion, shape and pseudopodial dynamics of unstimulated Dictyostelium cells are not random. J Cell Sci. 1993;106: 1005–1013.
17. Weiner OD, Marganski WA, Wu LF, Altschuler SJ, Kirschner MW. An Actin-Based Wave Generator Organizes Cell Motility. PLoS Biol. 2007;5: e221.
18. Graziano BR, Weiner OD. Self-organization of protrusions and polarity during eukaryotic chemotaxis. Curr Opin Cell Biol. 2014;30C: 60–67.
19. Miki H, Fukuda M, Nishida E, Takenawa T. Phosphorylation of WAVE downstream of mitogen-activated protein kinase signaling. J Biol Chem. 1999;274: 27605–27609.
20. Lebensohn AM, Kirschner MW. Activation of the WAVE complex by coincident signals controls actin assembly. Mol Cell. 2009;36: 512–524.
21. Machesky LM, Insall RH. Scar1 and the related Wiskott-Aldrich syndrome protein, WASP, regulate the actin cytoskeleton through the Arp2/3 complex. Curr Biol. 1998;8: 1347–1356.
22. Clark-Lewis I, Sanghera JS, Pelech SL. Definition of a consensus sequence for peptide substrate recognition by p44mpk, the meiosis-activated myelin basic protein kinase. Journal of Biological Chemistry. 1991;266: 15180–15184.
23. Eden S, Rohatgi R, Podtelejnikov AV, Mann M, Kirschner MW. Mechanism of regulation of WAVE1-induced actin nucleation by Rac1 and Nck. Nature. 2002;418: 790–793.
24. Kunda P, Craig G, Dominguez V, Baum B. Abi, Sra1, and Kette control the stability and localization of SCAR/WAVE to regulate the formation of actin-based protrusions. Curr Biol. 2003;13: 1867–1875.
25. Weiner OD, Rentel MC, Ott A, Brown GE, Jedrychowski M, Yaffe MB, et al. Hem-1 complexes are essential for Rac activation, actin polymerization, and myosin regulation during neutrophil chemotaxis. PLoS Biol. 2006;4: e38.
40
26. Koronakis V, Hume PJ, Humphreys D, Liu T, Horning O, Jensen ON, et al. WAVE regulatory complex activation by cooperating GTPases Arf and Rac1. Proc Natl Acad Sci U S A. 2011;108: 14449–14454.
27. Millius A, Watanabe N, Weiner OD. Diffusion, capture and recycling of SCAR/WAVE and Arp2/3 complexes observed in cells by single-molecule imaging. J Cell Sci. 2012;125: 1165–1176.
28. Faix J, Weber I. A dual role model for active Rac1 in cell migration. Small GTPases. 2013;4: 110–115.
29. Shutes A, Onesto C, Picard V, Leblond B, Schweighoffer F, Der CJ. Specificity and mechanism of action of EHT 1864, a novel small molecule inhibitor of Rac family small GTPases. J Biol Chem. 2007;282: 35666–35678.
30. Amato C, Thomason P, Davidson A, Swaminathan K, Ismail S, Machesky L, et al. WASP restricts active Rac to maintain cells’ front-rear polarisation. 2019
31. Kumar R, Sanawar R, Li X, Li F. Structure, biochemistry, and biology of PAK kinases. Gene. 2017;605: 20–31.
32. Coué M, Brenner SL, Spector I, Korn ED. Inhibition of actin polymerization by latrunculin A. FEBS Lett. 1987;213: 316–318.
33. Gerisch G, Bretschneider T, Muller-Taubenberger A, Simmeth E, Ecke M, Diez S, et al. Mobile actin clusters and traveling waves in cells recovering from actin depolymerization. Biophys J. 2004;87: 3493–3503.
34. Millius A, Dandekar SN, Houk AR, Weiner OD. Neutrophils establish rapid and robust WAVE complex polarity in an actin-dependent fashion. Curr Biol. 2009;19: 253–259.
35. McRobbie SJ, Newell PC. Chemoattractant-mediated changes in cytoskeletal actin of cellular slime moulds. J Cell Sci. 1984;68: 139–151.
36. Insall RH, Borleis J, Devreotes PN. The aimless RasGEF is required for processing of chemotactic signals through G-protein-coupled receptors in Dictyostelium. Curr Biol. 1996;6: 719–729.
37. Wu L, Valkema R, Van Haastert PJ, Devreotes PN. The G protein beta subunit is essential for multiple responses to chemoattractants in Dictyostelium. J Cell Biol. 1995;129: 1667–1675.
38. Maeda M, Aubry L, Insall R, Gaskins C, Devreotes PN, Firtel RA. Seven helix chemoattractant receptors transiently stimulate mitogen-activated protein kinase in Dictyostelium. Role of heterotrimeric G proteins. J Biol Chem. 1996;271: 3351–3354.
41
39. Peracino B, Borleis J, Jin T, Westphal M, Schwartz JM, Wu L, et al. G protein beta subunit-null mutants are impaired in phagocytosis and chemotaxis due to inappropriate regulation of the actin cytoskeleton. J Cell Biol. 1998;141: 1529–1537.
40. Nichols JME, Paschke P, Peak-Chew S, Williams TD, Tweedy L, Skehel M, et al. The Atypical MAP Kinase ErkB Transmits Distinct Chemotactic Signals through a Core Signaling Module. Developmental Cell. 2019
41. Scherle PA, Jones EA, Favata MF, Daulerio AJ, Covington MB, Nurnberg SA, et al. Inhibition of MAP kinase kinase prevents cytokine and prostaglandin E2 production in lipopolysaccharide-stimulated monocytes. J Immunol. 1998;161: 5681–5686.
42. Niewohner J, Weber I, Maniak M, Muller-Taubenberger A, Gerisch G. Talin-null cells of Dictyostelium are strongly defective in adhesion to particle and substrate surfaces and slightly impaired in cytokinesis. J Cell Biol. 1997;138: 349–361.
43. Simson R, Wallraff E, Faix J, Niewöhner J, Gerisch G, Sackmann E. Membrane bending modulus and adhesion energy of wild-type and mutant cells of Dictyostelium lacking talin or cortexillins. Biophysical journal. 1998;74: 514–522.
44. Tsujioka M, Yoshida K, Nagasaki A, Yonemura S, Muller-Taubenberger A, Uyeda TQ. Overlapping functions of the two talin homologues in Dictyostelium. Eukaryot Cell. 2008;7: 906–916.
45. Davidson AJ, Amato C, Thomason PA, Insall RH. WASP family proteins and formins compete in pseudopod- and bleb-based migration. J Cell Biol. 2018;217: 701–714.
46. Blagg SL, Stewart M, Sambles C, Insall RH. PIR121 regulates pseudopod dynamics and SCAR activity in Dictyostelium. Curr Biol. 2003;13: 1480–1487.
47. Chubb JR, Wilkins A, Thomas GM, Insall RH. The Dictyostelium RasS protein is required for macropinocytosis, phagocytosis and the control of cell movement. J Cell Sci. 2000;113: 709–719.
48. Andrew N, Insall RH. Chemotaxis in shallow gradients is mediated independently of PtdIns 3-kinase by biased choices between random protrusions. Nat Cell Biol. 2007;9: 193–200.
49. Bosgraaf L, Van Haastert PJ. Navigation of chemotactic cells by parallel signaling to pseudopod persistence and orientation. PLoS One. 2009;4: e6842.
50. Suraneni P, Rubinstein B, Unruh JR, Durnin M, Hanein D, Li R. The Arp2/3 complex is required for lamellipodia extension and directional fibroblast cell migration. J Cell Biol. 2012;197: 239–251.
42
51. Giannone G, Dubin-Thaler BJ, Rossier O, Cai Y, Chaga O, Jiang G, et al. Lamellipodial actin mechanically links myosin activity with adhesion-site formation. Cell. 2007;128: 561–575.
52. Insall RH. Understanding eukaryotic chemotaxis: a pseudopod-centred view. Nat Rev Mol Cell Biol. 2010;11: 453–458.
53. Humphries JD, Paul NR, Humphries MJ, Morgan MR. Emerging properties of adhesion complexes: what are they and what do they do. Trends Cell Biol. 2015;25: 388–397.
54. Hannezo E, Heisenberg C-P. Mechanochemical Feedback Loops in Development and Disease. Cell. 2019;178: 12–25.
55. Davidson AJ, King JS, Insall RH. The use of streptavidin conjugates as immunoblot loading controls and mitochondrial markers for use with Dictyostelium discoideum. Biotechniques. 2013;55: 39–41.
Tubulin
Phosphatase - + - +
NIH
3T3
NIH
3T3
MD
A-M
B-2
31
MD
A-M
B-2
31
WAVE2
ScarMccc1
Sca
r-
A B C
E D
Phosphatase - +
WT
WT
Sca
r-
*****
Figure 1 Singh et al.
***
0.0 0.1 0.2 0.3 0.4
50
100
150
200
250
Distance
Inte
nsity
(A.U
.)
0.0 0.2 0.4 0.6
100
200
300
Distance 0.0 0.2 0.4
100
200
300
Distance
-Phosphatase+Phosphatase
0.05 0.1 0.15 0.2 0.25
100
200
300
Distance
MDA-MB-231NIH3T3
Inte
nsity
(A.U
.)
Inte
nsity
(A.U
.)
Inte
nsity
(A.U
.) MDA-MB-231
0.05 0.1 0.15 0.2 0.25 0.350
100
150
200
250
Distance
NIH3T3-Phosphatase+Phosphatase
0.0 0.1 0.2 0.3
50
100
150
200
250
Distance
-Phosphatase+Phosphatase
Inte
nsity
(A.U
.)
WAVE2
Tubulin
B 1 255-256 376-377 443
SHD Poly-proline V C A
F
S S S S S SS S S S SS S
287 290 301 335 339 384 388 389 430 431 432 437 440
Y Y Y
88 129 210
S8S13
S5S3
Dd Scar
G HIn Scar-
IIn Scar-
Sca
rWT
Sca
rY88
F
Sca
rY88
/129
F
Sca
rY88
/129
/210
F
Sca
rS3A
Sca
rS5A
Sca
rS5D
Sca
r-
Sca
rWT
Sca
rdVC
A
****
****Scar
Mccc1
In Scar-
J
In Scar-
Sca
rWT
Sca
rS8A
Sca
rS8D
*
K
Sca
rS13
D
Sca
rS13
A
In Scar-
Scar
Mccc1
In Scar-
- + - + - + Phosphatase
Sca
rS8A
Sca
rS8D
Sca
rWT
B 1 170 218 411 498
SHD Poly-proline V C A
S S S S S S S T S S S S S S
293 296 298 308 343 346 351 429 442 482 484 488 489 497
S8/T1
WAV
E1/
2 K
O
WAV
E2W
T
WAV
E2S
8A/T
1A
WAVE2
Tubulin
M
N
Hs WAVE2
ScarMccc1
L
Pak
-CR
IB EHT1864 (50 μM) A B
C
I
Scar
Mccc1
PIR
121-
WT
A-
site
WT
A s
ite
PIR
121G
fp in
Pir1
21-
Figure 2 Singh et al.
F
K
Latrunculin A
0 10 20 30 min
5 min 10 min 15 min
19 min
Sca
r com
plex
J
L
EHT1864 (50 μM)
-15s 60s
Pak-CRIB Scar Complex Merged
Scar
Mccc1
Ax3
LatrunculinA
**** *
0 h 6 h 0 s 90 s
Scar
Mccc1
0 2 5 10 min EHT1864
EHT1864 0 2 4 6 h
WAVE2
***** *
**
pERK
ERK 0 2 4 6
0.0
0.5
1.0
1.5
2.0
Time (h)
Inte
nsity
(upp
er/lo
wes
t)
0.1 0.2 0.3
50
100
150
200
250
Distance
Inte
nsity
(A.U
.)
0246
0.0 0.1 0.2 0.3 0.4
100
200
300
Distance
Inte
nsity
(A.U
.)
02510
D E
G H
0 2 5 100
20
40
60
Time (min)
% P
hosp
hory
latio
n
19 min
A B
Mccc1
Scar
Folate 0 20 40 60 120 240 Sec
Figure 3 Singh et al.
C
Scar
Mccc1
Sca
r-
Erk
A-
Erk
B-
Erk
AB
-
WT
WAVE2
pErk
Erk
Uo126 0 5 10 20 30 60
Mccc1
Scar
WAVE2
pErk
Erk
Uo126 0 5 10 20 30 60
In Dictyostelium
B16F1
NIH3T3
D EMccc1
pERK
pERK
Mccc1
cAMP 0 20 40 60 120 240 sec
0 20 40 60 120 2400
10
20
30
40
50
Time (Sec.)
% P
hosp
hory
latio
n
Scar (Folate,50 μM)
0 20 40 60 120 2400
2
4
6
Time (Sec.)
Rel
ativ
e fo
ld c
hang
e
pErkB (Folate, 50 μM)
0 20 40 60 120 2400
10
20
30
Time (Sec.)
% P
hosp
hory
latio
n
Scar (cAMP, 1μM)
0
5
10
15
20 pERKB (cAMP, 1μM)
Time (Sec.)
Rel
ativ
e fo
ld c
hang
e
0 20 40 60 120 240
Scar
Mccc1
WT
Gβ−
WT
15 min
Gβ-0
10
20
30
40
% P
hosp
hory
latio
n
0.05 0.15 0.25 0.35 0.4550
100
150
200
250
300
Distance
Inte
nsity
(A.U
.) WTErkA-ErkB-ErkAB-
0 5 10 20 30 600.0
0.5
1.0
1.5
2.0
Time (min)
Inte
nsity
(Top
/bot
tom
)
0 5 10 20 30 600.0
0.5
1.0
1.5
2.0
2.5
Time (min)
Inte
nsity
(Top
/bot
tom
)
F
B16F1
NIH3T3
G H I J
K L
M N
a
b
A
FE
G
L
N
WAVE2
Tubulin
JV
M3
Jur
kat
NIH
3T3
PD
AC
A
PD
AC
B
Cos
7
MD
A-M
B-2
31
B16
F1
Figure 4 Singh et al.
0 2 5 10 20 30
Time after adhesion (min)
Mccc1
Scar *******
a
b
0 1 2 5 10 15
Scar
Mccc1
Time after deadhesion
***** *
0 2 5 10 20 300
10
20
30
40
50
Adhesion (Time, min)
% P
hosp
hory
latio
n0 1 2 5 10 15
0
10
20
30
40
50
Suspension (Time, min)%
Pho
spho
ryla
tion
0.0 0.1 0.2 0.3
100
200
300
Distance
Vegetative
Suspension Adhesion
Inte
nsity
(A.U
.)
25 50 75 100 Confluency (%)
WAVE2
Tubulin
WAVE2
Tubulin
B16F1
NIH3T3
25 50 75 1000.0
0.5
1.0
1.5
2.0
Confluency (%)
Inte
nsity
(Upp
er/lo
wes
t) B16F1
25 50 75 1000.0
0.5
1.0
1.5
2.0
2.5
Confluency (%)
Inte
nsity
(Upp
er/lo
wes
t) NIH3T3
JVM3
Jurka
t
NIH3T
3
PDAC-A
PDAC-B
COS-7
MDA-MB-23
1
B16F1
0
1
2
3
Inte
nsity
(Upp
er/lo
wes
t)
*
Suspension Adhesion
Scar
Mccc1
WT
Talin
A/B
-
Suspension
WT
Talin
A/B
-
WT
adhe
sion
Talin
A/B
- fo
rced
adh
esio
n
Petri Plate
H
Distance
Suspension
0.0 0.1 0.2 0.30
20
40
60
Inte
nsity
(A.U
.) WTTalinA-/B-
Mccc1
Scar
Sus
pens
ion
Adh
esio
n
******
D
B
C
Distance
Plate
0.0 0.1 0.2 0.30
10
20
30
40
WTTalinA-/B-
I
Distance
Forced adhesion
0.2 0.4 0.6
20406080
100WTTalinA-/B-
J
K
M O
B
D
************
*
******
**** ****ns
0 15 30 45 60 s S
carW
T S
carS8
ASc
arS8
D
In S
car-
Sca
r-
AIn
Sca
r-
C
Figure 5 Singh et al.
0 6 18 24s 12
E
F G
Scar-0
5
10
15
20202530
Spee
d (μ
m/m
in)
Scar-0.0
0.2
0.4
0.6
0.8
1.0
1.2
Tortu
osity
ns
********
********
****
********
*****ns
ns
0
5
10
15
20
Acc
umul
atio
n si
ze (μ
m)
0
5
10
15
20
Patc
h /m
in
F-A
ctin
WAVE1/2 KO WAVE2WT WAVE2S8A/T1AH
0
25
Patc
h lif
e tim
e (S
ec)
Sca
rWT
Sca
rS8A
Scar
S8D
ScarWT ScarS8A ScarS8D
ScarWT ScarS8A ScarS8D
ScarWT ScarS8A ScarS8D
ScarWT ScarS8A ScarS8D ScarWT ScarS8A ScarS8D
Scar
Mccc1
Doxycycline + - + - + - + -
In Scar-/WASP-
A B
0 15 30 45 60 s
In S
car-
/WA
SP-
Sca
r-/W
ASP
-
In Scar-/WASP-
Scar
com
plex
E
F- A
ctin
In Scar-/WASP-
Figure 6 Singh et al
C D
F
Scar-/WASP-
******** ****
0
4
8
12
16
Spee
d
0 24 48 72 96 120 1440
10
20
30
40
50
Time (h)
Cel
l no.
(x10
5 )
Scartet-/WASP-
WT
S8A
S8D
Sca
rWT
Sca
rS8A
Scar
S8D
ScarWT ScarS8A ScarS8D ScarWT ScarS8A ScarS8D
ScarWT ScarS8A ScarS8D
ScarWT ScarS8A ScarS8D
A
- CycloheximideWT S8AS8D
0 2 4 6 0 2 4 6 0 2 4 60.0
0.5
1.0
Time (h)
Rel
ativ
e fo
ld c
hang
e
+ Cycloheximide
B
FIgure 7 Singh et al
C
WT S8AS8D
0 2 4 6 0 2 4 6 0 2 4 60.0
0.5
1.0
1.5
Time (h)
Rel
ativ
e fo
ld c
hang
e
ScarWT
Mccc1
ScarS8A
Mccc1
ScarS8D
Mccc1
0 2 4 6 2 4 6 h
- Cycloheximide + Cycloheximide
0 2 5 10 20 300.0
0.5
1.0
Adhesion time (min)
Fold
cha
nge
(Tot
al S
car)
0 1 2 5 10 150
1.0
2.0
Suspension time (min)
Fold
cha
nge
(Tot
al S
car)
D E
Table 1 Phosphorylated Residues in Scar
Peptide Sequence Identification Source
Y88 PSIEDYHRNTS By MS and Screening
Y129 SINTVYEKCKP By MS and Screening
Y210 VTKVRYDPVTG By MS and Screening
S287/S290 PPLNTSTPSPSSSF By MS and Screening
S301 QGRPPSTGFNT By MS and Screening
S335/S339 ANNRLSVHNSAPIVA Screening S384/S388/S389 ASGARSDLLSSIMQGM By MS and
screening Phosphorylated Residues in WAVE2 S293/S296/S298 GPKRSSVVSPSHPPPA Chen et al., 2010,
Lebensohn and Kirschner 2009, Phosphosite.org,
S308 APPLGSPPGPK Lebensohn and Kirschner 2009
S343/T346/S351 PVGFGSPGTPPPPSPPSFP Mendoza et al., 2011, Danson et al., 2007
S429 TKPKSSLPAVS Mertins et al., 2016, Phosphosite.org.
S442 RSDLLSAIRQG Mertins et al., 2016, Phosphosite.org.
Supp table 1; supp figs 1-3
A
B
Suppl. figure 1 Singh et al.
50
80
120
140
Scar
EGFP-NAP1PIR121
Scar-/EGFPNap1 EGFPNap1
50
80
140
Scar-/EGFPNap1 EGFPNap1
InputSup Pulldown PulldownSupInput
PulldownSupInputPulldownInputSup
1 2 3 4 5 6
C
Scar
Mccc1
In Scar-
Sca
rWT
Sca
rS8A
Sca
rS8D
Sca
r
Sca
rWT
Sca
rS8A
Sca
rS8D
Scar
Pir121Nap1
Ub
Scar
Mccc1
Ub
Gfp Trap pull down Whole cell lysate- MG132 + MG132 + MG132- MG132
A
Suppl. Figure 2 Singh et al
B
Sca
r
Sca
rWT
Sca
rS8A
Sca
rS8D
Sca
r
Sca
rWT
Sca
rS8A
Sca
rS8D
Sca
r
Sca
rWT
Sca
rS8A
Sca
rS8D