During an animal’s lifetime, many of its cells will move from one location in the body to another. For example, skin cells can migrate to repair wounds. Prior to migration, cells are usually attached to a scaffold called the extracellular-matrix, which helps to hold them in a particular location within a tissue. Individual cells can move in different ways. During a type of movement called mesenchymal migration, the front end of a cell grows outwards and attaches to a different section of the matrix. The rear of the cell is pulled forward and it detaches from the matrix and retracts, which allows the entire cell to move forward. In contrast, during amoeboid migration, the moving cells are only loosely attached to the matrix and move by gliding.

There are large variations in how cells move and they can adopt modes that lie between the two extremes of mesenchymal and amoeboid migration. They can also switch between modes depending on their requirements. Shafqat-Abbasi et al. developed a method to analyse how individual human lung cancer cells move. The method uses software to collect data on cell shape, speed of movement and other features from microscopy images of the migrating cells.

The experiments reveal that the cells adopt two distinct migration modes, which Shafqat-Abbasi et al. termed 'Discontinuous' and 'Continuous'. The majority of cells migrated in the Discontinuous mode, in which cells moved in many different directions. This was caused by a lack of coordination between the outgrowth of the front end of the cell, and the retraction of the back from the matrix. In contrast, in the cells that migrated using the Continuous mode, an outgrowth consistently led to a retraction, which enabled cells to move in one direction.

Further experiments revealed that the mode of migration used by the cells is affected by how tightly they are bound to the extracellular-matrix, and the mechanical forces generated inside the cells to drive the movement. Shafqat-Abbasi et al.’s method provides an analytical toolbox that other researchers can use to study the mesenchymal migration of animal cells.

Cell migration is a profoundly heterogeneous phenomenon. Indeed, cells can adopt several substantially different migration modalities, including multicellular, amoeboid, and mesenchymal (also termed lamellipodial or lamellipodial-driven) migration, which can all be utilized by a broad range of cell types, as well as lobopodial migration, which has been observed specifically in fibroblasts (Friedl and Wolf, 2010; Sahai, 2005; Petrie and Yamada, 2015; Petrie et al., 2014; Welch, 2015; Friedl and Alexander, 2011). These migration modes represent 'prespecified' cellular configurations (i.e. cell states) that are favored under particular conditions (Friedl, 2004). Switch-like conversion between these distinct modes is therefore part of the plastic, adaptive/compensatory response of cells to either environmental modulation (Liu et al., 2015; Starke et al., 2014; Ruprecht et al., 2015) or molecular targeting (Sahai et al., 2007; Sanz-Moreno et al., 2008; Somlyo et al., 2003; Wolf, 2003). At a finer scale, heterogeneity is also evident within these migration modes, arising either stochastically or as an adaptive response to changing cues (Geiger et al., 2009; Lämmermann and Sixt, 2009; Lock et al., 2014; Winograd-Katz et al., 2009). Yet, partly due to a lack of adequate quantification, it remains unclear to what extent variation within modes occurs either progressively along a continuum or in a switch-like manner between as yet undefined intra-modal subpopulations. Specifically, in the case of amoeboid migration, three discrete sub-modalities have been observed, and these can co-exist under individual conditions (Welch, 2015; Lämmermann and Sixt, 2009; Yoshida, 2006). By contrast, potentially distinct styles of mesenchymal (lamellipodial) migration, including keratocyte-like (Barnhart et al., 2015; Keren et al., 2008) and fibroblast-like (Abercrombie et al., 1977; Theisen et al., 2012) migration, have been described as arising largely in separate cell types or conditions. Therefore, despite some early suggestions (Lewis et al., 1982), it has remained uncertain to what extent divergent sub-modalities of mesenchymal migration spontaneously emerge in parallel within uniform cell populations and conditions, and whether these modes are truly quantitatively distinct, or instead represent extremes in a broad phenotypic continuum. These questions are important because each discrete migration mode, with its unique internal logic and dependencies, may respond differently to altered regulatory cues – thereby defining a multifaceted adaptive response. Such responses are exemplified by the seminal observation that amoeboid motility provides a compensatory escape route for tumor cells when mesenchymal migration is blocked via inhibition of proteolytic mechanisms or by confinement (Wolf, 2003). Identifying and characterizing such distinct modes within individual cell types and/or conditions may therefore be key to understanding heterogeneity in response to experimental or even clinical interventions.

Unlike amoeboid migration, which requires little or no cell-matrix adhesion (Lämmermann et al., 2008), mesenchymal migration demands cell adhesion to the extracellular matrix (ECM) for effective force application (Friedl and Wolf, 2010). This adhesion is achieved through the interaction of integrin-mediated cell-matrix adhesion complexes (CMACs) with ECM ligands, while applied forces are generated by the F-actin cytoskeleton and actomyosin system (Geiger et al., 2009; Humphries et al., 2015). This arrangement defines the ECM – adhesion – F-actin axis that directly mediates mesenchymal migration (Parsons et al., 2010; Schwartz, 2010).

Integrins are central to the ECM – adhesion – F-actin axis and are both anchored and activated by binding through their extracellular domains to ECM ligands, such as collagen, laminin, and fibronectin (Lewis et al., 1982). The composition, topology, and density of ECM ligands therefore play key roles in mesenchymal migration (Geiger et al., 2009). Through interaction with integrin tails, talin acts a critical intracellular regulator of integrin activation (Calderwood et al., 1999; Kiss et al., 2015; Moser et al., 2009; Tadokoro, 2003), and is also one of several CMAC proteins linking integrins to the F-actin cytoskeleton (Schwarz and Gardel, 2012). This enables cellular force application from F-actin, through CMACs, to the ECM, thereby driving membrane protrusion/retraction and cell translocation (Small and Resch, 2005). Furthermore, CMACs function both up- and downstream of the small GTPases Rac1 and RhoA (Raftopoulou and Hall, 2004), which modulate both F-actin polymerization and actomyosin contractility, the latter through the Rho kinase (ROCK)-mediated regulation of non-muscle myosin II (Ridley, 2003; Riento and Ridley, 2003). In fact, CMACs function as hubs for bi-directional chemical and mechanical information transduction across the plasma membrane, providing command and control of mesenchymal migration (Hynes, 2002; Lock et al., 2008).

Given the pivotal roles of CMACs and F-actin, their assessment can efficiently provide a broad estimation of how the complex machinery underlying mesenchymal migration is organized (Lock et al., 2014; Gardel et al., 2010; Gupton and Waterman-Storer, 2006; Kim and Wirtz, 2013). Such an approach, generally based on quantitative imaging, is especially effective when multiscale data capturing both cell behavior (migration) and organization (e.g. CMAC and F-actin status) is derived simultaneously on a per cell basis. This facilitates the leveraging of natural or induced heterogeneity to define: i) the statistical structure of variation (e.g. progressive or discrete) within and between cell populations, as well as; ii) key trends, dependencies and relationships within the cell migration system (Lock et al., 2014; Keren et al., 2008; Kiss et al., 2015; Kowalewski et al., 2015; Ku et al., 2012; Lee et al., 2015; Lock and Strömblad, 2010). Here, we extend on this approach by employing a tailored analytical toolbox to detect two quantitatively distinct sub-modalities of mesenchymal migration, thus illuminating the discrete nature of variation within this general migration archetype. We employ this unique suite of analytical tools to further characterize key aspects of the behavior, organization, and regulation of these divergent migration strategies. Collectively, this study now provides conceptual and practical capabilities to the cell migration research community, while also highlighting the importance of distinguishing individual mesenchymal migration modes as a vital precursor to understanding mesenchymal migration as a whole.

We here present an integrated analytical approach and associated tools, designed to interrogate both spontaneous and induced heterogeneity in single-cell mesenchymal (lamellipodial) migration. By deploying these tools, we quantitatively demonstrate that mesenchymal migration is composed of two distinct sub-modalities, each with specific behavioral, organizational, and regulatory characteristics. This marks an important advance that may enhance future investigations of the mesenchymal migration archetype. In particular, interpretations of natural and/or experimentally induced variability will likely become more coherent and precise if data are disaggregated according to Discontinuous or Continuous modes, rather than being unintentionally aggregated across cells with fundamentally different properties and dependencies. Recognition of these modes may therefore facilitate a more accurate understanding of the heterogeneity, adaptability, and regulation of mesenchymal migration as a whole.

Our initial classification of cells into Discontinuous and Continuous migration modes was based on observed behavioral differences. Crucially, the same distinctive behaviors emerged in parallel within a variety of cell lines and under a spectrum of conditions, thus supporting the robustness and broad relevance of these behaviors. Yet, given the similarity of Continuous and Discontinuous migrations to previous descriptions of keratocyte-like (Barnhart et al., 2015; Keren et al., 2008) or fibroblast-like motility (Abercrombie et al., 1977; Theisen et al., 2012), respectively, the existence of these behaviors per se is not novel. Indeed, early observations even qualitatively suggest the co-existence of similar behaviors (Lewis et al., 1982). However, we now present quantitative evidence of the spontaneous, parallel emergence of two distinct yet inter-convertible mesenchymal migration modes, wherein divergent couplings exist between behavioral (speed) and organizational characteristics. We thereby provide new insights into the discontinuous structure of heterogeneity within the broad mesenchymal migration archetype. Indeed, by differentiating between progressive (intra-modal) and discrete (inter-modal) forms of variation, we gained the capacity to compare the roles that these two mechanisms play in adaptive responses to perturbation. Surprisingly, this revealed that adaptive switching between modes is the dominant response to perturbation (FN concentration modulation), with adaptive stretching (remodelling) within modes playing only a minor role.

The quantitative distinction between Discontinuous and Continuous migration modes is based on systemic differences detected via multivariate clustering of data following visual classification. Importantly, given that cells were classified based on perceived differences in their behavioral dynamics, we excluded behavioral features and focused only on organizational features (predominantly related to CMAC and F-actin status) when addressing the question of separation between the modes. As a result, at least four lines of evidence support the discrete nature of Discontinuous and Continuous migration modes.

First, supervised multivariate clustering (canonical vectors analysis, CVA) of mode-classified cell observations indicated near complete separation between Discontinuous and Continuous modes. The systemic organization underlying these migration modes is therefore, to a very large extent, quantitatively distinct.

Second, although separation between modes was not complete when cell observations were clustered in unsupervised PCA space, the topology of probability-density distributions was consistent with the existence of subpopulations in cell state space. Precisely, the high probability-density 'valleys' near the center of mass of each mode contrast with the low probability-density 'ridges' at the boundaries between modes. Importantly, these probability-density distributions are not predetermined outcomes of such PCA clustering, since PCA is unsupervised with respect to mode identity. In fact, these topologies are suggestive of a system comprised of two attractor states (Huang and Ingber, 2000), whose basins of attraction diverge on either side of a set of potentially unstable or unfavorable configurations. The low probability of these boundary configurations effectively divides mesenchymal migration into two independent sub-modalities. Such attractor states have been described in relation to, for example: gene expression during development and tumorigenesis (Huang et al., 2009; Huang, 2009); cell signaling during motility (Kim et al., 2015); and even in the force-coupled dynamics of CMACs (Hernández-Varas et al., 2015). Moreover, the attractor state concept appears analogous to the concept of prespecification originally proposed by Friedl to explain the robust, recurrent and yet switchable characteristics of mesenchymal and amoeboid migration modes (Friedl, 2004). Interestingly, the limited remodelling within Discontinuous and Continuous modes in response to FN modulation (as opposed to mode switching) suggests that these modes may be relatively inflexible, providing additional support for their proposed status as prespecified attractor states.

Third, further indicating the discrete nature of Discontinuous and Continuous migration modes, CVA clustered cell observations grouped by both migration mode and cell speed were again largely distinct, supporting the validity of the original CVA clustering result. In addition, this approach also provided a fourth compelling line of evidence by revealing completely independent trajectories through cell state space defined by cells migrating at different speeds within each mode. Remarkably, this indicates that changes in cell speed within each mode were coupled to substantially different patterns of organizational remodelling. These systemic differences were composed in part by specific disparities in Spearman’s correlations between individual organizational features and cell speed within each mode. While a variety of differences were observed, most striking were instances where correlations to cell speed were inverted between modes, such as in the case of adhesion stability (CMAC lifetime) – a critical parameter previously linked to causal regulation of cell speed (Lock et al., 2014). Interestingly, when comparing modes, the independent patterns of cell speed-coupled reorganization arose despite only subtle differences across the entire network of inter-feature Spearman’s correlations. Indeed, these differences in correlation values were far more limited than would be expected at random, suggesting that a select few features and relationships may control migration mode identity.

Taking into account each of these lines of evidence, we conclude that Discontinuous and Continuous migration modes are representative of more than simply cells with different organizational feature values. We propose that these modes in fact represent divergent, yet co-existing strategies of mesenchymal migration.

Despite strongly embracing quantitative imaging-based research (Lock and Strömblad, 2010; Le Dévédec et al., 2010; Masuzzo and Martens, 2015), the cell migration field has seen remarkably little application of automated approaches to the classification of migrating cells. This is in contrast to the study of mitosis, where automated classification has already proven to be a powerful research tool (Neumann et al., 2010; Schmitz et al., 2010). It is likely that this reflects differences in the degree and structure of heterogeneity in these biological processes. For example, mitosis is a relatively stereotyped process - both morphologically and particularly in the terms of the temporal ordering of events. Both characteristics can increase the accuracy of automated classification approaches (Held et al., 2010). While cell migration is also more (e.g. Discontinuous mode) or less (e.g. Continuous mode) a globally time-ordered process, the underlying events appear far less stereotypical. Given these challenges, we have employed a blinded manual classification approach in this study. However, we note that the CVA clustering results – even excluding behavioral features (as herein) – provide evidence that automated classification of migration modes may be feasible in future. This may significantly enhance subsequent quantitative analyses of mode organization, behavior, and regulation.

Such automated classification may be in part based on our detailed spatiotemporal analysis of protrusive and retractive membrane dynamics within each mode. This revealed that the major differences between modes were: the size of membrane retraction events; the order and temporal coupling of protrusion and retraction events; and the coordination displayed between protrusive and retractive dynamics. Specifically, cells in the Discontinuous mode produced many very small and occasional very large retraction events, with these retractions tending to lead protrusion, reversing the order detected in Continuous cells. A large time delay also exists between retraction and subsequent protrusion dynamics in Discontinuous cells, and these dynamics are less coordinated (lower cross-correlation). Such substantial differences in membrane dynamics (especially the inversion of protrusion and retraction orders) further support the distinct nature of Discontinuous and Continuous migration modes. Furthermore, these analyses both confirmed and explained the visual impressions that originally differentiated Discontinuous from Continuous migration. In addition, given the strong correspondence between retraction (but not protrusion) size and cell speed, these analyses now add migration mode and cell speed to motility initiation as aspects of migration where membrane retraction appears to be the deterministic membrane process (Barnhart et al., 2015; Kowalewski et al., 2015; Cramer, 2010).

Having quantitatively established the discrete nature and behavioral characteristics of Discontinuous and Continuous migration modes, we sought to understand how their frequency balance is regulated. Importantly, to select candidate molecular regulators, we first considered the differences in organizational feature values that arose spontaneously between the modes. Then, given existing knowledge on how targeting of particular molecular components (FN concentration [Gupton and Waterman-Storer, 2006]; talin 1 expression [Kiss et al., 2015]; ROCK signaling [Lock et al., 2014; Kim and Wirtz, 2013; Hernández-Varas et al., 2015]) impact these same features, we designed a series of perturbations intended to drive specific changes in these features, while also monitoring the behavioral balance between modes. This demonstrated that each of these molecular factors plays a key role in shaping the equilibrium between modes. Furthermore, these molecular targets also regulate generic characteristics of the pivotal ECM – adhesion – F-actin axis, including: integrin ligation (FN concentration); integrin activation and clustering (talin expression); integrin – F-actin linkage (talin expression); and actomyosin contractility (ROCK signaling). Hence, the effects of these specific perturbations may also be illustrative of how generalized regulatory mechanisms impact the migration mode equilibrium.

Taking this perspective, we observe that while Continuous migration arises under conditions of high adhesion and full contractility, Discontinuous migration is preferred given low adhesion and inhibited contractility. It is noteworthy that both modes emerge under conditions that are distinct from those that favour amoeboid migration (i.e. low adhesion and high contractility). Therefore, the switch between Continuous and Discontinuous (mesenchymal) migration modes clearly contrasts to the previously identified conversion between mesenchymal and amoeboid migration. This emphasizes the multifaceted adaptive capacity of migrating cells, which likely expands the contexts within which efficient migration is possible. This plasticity is highlighted by the growing variety of distinct migration modalities now recognized (Figure 7G).

Finally, by viewing differences in organizational feature values (both spontaneous and perturbation-induced) and mode frequencies as a logical matrix, we could identify organizational features that are coupled to migration mode identity under all observed conditions, as well as features that can be explicitly uncoupled. This provides a first step in parsing correlated features from those with significant causal influence over migration mode identity. Specifically, it is notable that CMAC Area and CMAC Lifetime were previously indicated to causally influence cell migration speed (Lock et al., 2014; Kim and Wirtz, 2013), while Cell Compactness was causally downstream of cell speed (Lock et al., 2014). Thus, the evidence that these features may be functionally coupled to the determination of migration mode identity is consistent with them being integral to migratory regulation. Conversely, we previously found that Cell Area showed no causal relationship to cell speed (Lock et al., 2014), and this is again consistent with it being functionally uncoupled from the process of mode identity determination. Overall, three types of insight were revealed through these perturbation-based analyses, including: the regulatory roles of specific molecules (fibronectin, talin and ROCK); the putative influence of generic regulatory mechanisms; and the nature of links (coupled vs uncoupled) between migration mode identity and commonly measured macromolecular features. Nonetheless, much remains to be learned about the proximal and distal determinants of these migration modes.

In conclusion, we have presented and applied a quantitative, imaging-based analytic approach to explore the heterogeneity that naturally emerges within the mesenchymal (lamellipodial) archetype of cell migration. We thereby characterized two quantitatively distinct migration modes, here termed Continuous and Discontinuous migrations, that co-exist within the broad mesenchymal migration archetype. We compared these migration modes in terms of: their motion and behavioral dynamics; the specific and systemic organization of key machineries (CMACs and F-actin) underlying their motility, and; differences in how behavior (migration speed) and organization co-evolve within each mode. Through targeted molecular perturbations, we defined both specific molecular and general mechanisms of bottom-up control over migration mode determination, while also beginning to parse potentially functional (coupled) relationships (between features and modes) from those that are correlated but non-functional (uncoupled). Finally, our systemic analysis of adaptive responses to perturbation has revealed how discrete mechanisms (inter-modal adaptive switching) dominate over progressive remodeling (intra-modal adaptive stretching). This study therefore emphasize the importance of distinguishing and comprehending Discontinuous and Continuous migration modes as a necessary precursor to understanding mesenchymal migration in its totality, while simultaneously providing tools and approaches that enable this endeavor.

1. 1. Shafqat-Abbasi H
   2. Kowalewski JM
   3. Kiss A
   4. Gong X
   5. Hernandez-Varas P
   6. Berge U
   7. Jafari-Mamaghani M
   8. Lock JG
   9. Strömblad S

   (2015) Data from: An analysis toolbox to explore mesenchymal migration heterogeneity reveals adaptive switching between distinct modes

   Available at Dryad Digital Repository under a CC0 Public Domain Dedication.

   http://dx.doi.org/10.5061/dryad.9jh6m

1. 1. Abercrombie M
   2. Heaysman JEM
   3. Pegrum SM

   (1970) The locomotion of fibroblasts in culture i. movements of the leading edge

   Experimental Cell Research 59:393–398.

   https://doi.org/10.1016/0014-4827(70)90646-4

   • Google Scholar
2. 1. Abercrombie M
   2. Dunn GA
   3. Heath JP

   (1977)

   The shape and movement of fibroblasts in culture

   Society of General Physiologists Series 32:57–70.

   • Google Scholar
3. 1. Barnhart E
   2. Lee K-C
   3. Allen GM
   4. Theriot JA
   5. Mogilner A

   (2015) Balance between cell−substrate adhesion and myosin contraction determines the frequency of motility initiation in fish keratocytes

   Proceedings of the National Academy of Sciences of the United States of America 112:5045–5050.

   https://doi.org/10.1073/pnas.1417257112

   • Google Scholar
4. 1. Calderwood DA
   2. Zent R
   3. Grant R
   4. Rees DJG
   5. Hynes RO
   6. Ginsberg MH

   (1999) The talin head domain binds to integrin subunit cytoplasmic tails and regulates integrin activation

   Journal of Biological Chemistry 274:28071–28074.

   https://doi.org/10.1074/jbc.274.40.28071

   • Google Scholar
5. 1. Chen WT

   (1981) Mechanism of retraction of the trailing edge during fibroblast movement

   The Journal of Cell Biology 90:187–200.

   https://doi.org/10.1083/jcb.90.1.187

   • Google Scholar
6. 1. Cramer LP

   (2010) Forming the cell rear first: breaking cell symmetry to trigger directed cell migration

   Nature Cell Biology 12:628–632.

   https://doi.org/10.1038/ncb0710-628

   • Google Scholar
7. 1. Friedl P

   (2004) Prespecification and plasticity: shifting mechanisms of cell migration

   Current Opinion in Cell Biology 16:14–23.

   https://doi.org/10.1016/j.ceb.2003.11.001

   • Google Scholar
8. 1. Friedl P
   2. Wolf K

   (2010) Plasticity of cell migration: a multiscale tuning model

   The Journal of Cell Biology 188:11–19.

   https://doi.org/10.1083/jcb.200909003

   • Google Scholar
9. 1. Friedl P
   2. Alexander S

   (2011) Cancer invasion and the microenvironment: plasticity and reciprocity

   Cell 147:992–1009.

   https://doi.org/10.1016/j.cell.2011.11.016

   • Google Scholar
10. 1. Gardel ML
    2. Schneider IC
    3. Aratyn-Schaus, Y
    4. Waterman CM

    (2010) Mechanical integration of actin and adhesion dynamics in cell migration

    Annual Review of Cell and Developmental Biology 26:315–333.

    https://doi.org/10.1146/annurev.cellbio.011209.122036

    • Google Scholar
11. 1. Geiger B
    2. Spatz JP
    3. Bershadsky AD

    (2009) Environmental sensing through focal adhesions

    Nature Reviews Molecular Cell Biology 10:21–33.

    https://doi.org/10.1038/nrm2593

    • Google Scholar
12. 1. Gupton SL
    2. Waterman-Storer CM

    (2006) Spatiotemporal feedback between actomyosin and focal-adhesion systems optimizes rapid cell migration

    Cell 125:1361–1374.

    https://doi.org/10.1016/j.cell.2006.05.029

    • Google Scholar
13. 1. Held M
    2. Schmitz MHA
    3. Fischer B
    4. Walter T
    5. Neumann B
    6. Olma MH
    7. Peter M
    8. Ellenberg J
    9. Gerlich DW

    (2010) CellCognition: time-resolved phenotype annotation in high-throughput live cell imaging

    Nature Methods 7:747–754.

    https://doi.org/10.1038/nmeth.1486

    • Google Scholar
14. 1. Hernández-Varas P
    2. Berge U
    3. Lock JG
    4. Strömblad S

    (2015) A plastic relationship between vinculin-mediated tension and adhesion complex area defines adhesion size and lifetime

    Nature Communications 6:7524.

    https://doi.org/10.1038/ncomms8524

    • Google Scholar
15. 1. Huang S
    2. Ingber DE

    (2000) Shape-dependent control of cell growth, differentiation, and apoptosis: switching between attractors in cell regulatory networks

    Experimental Cell Research 261:91–103.

    https://doi.org/10.1006/excr.2000.5044

    • Google Scholar
16. 1. Huang S
    2. Ernberg I
    3. Kauffman S

    (2009) Cancer attractors: a systems view of tumors from a gene network dynamics and developmental perspective

    Seminars in Cell & Developmental Biology 20:869–876.

    https://doi.org/10.1016/j.semcdb.2009.07.003

    • Google Scholar
17. 1. Huang S

    (2009) Reprogramming cell fates: reconciling rarity with robustness

    BioEssays 31:546–560.

    https://doi.org/10.1002/bies.200800189

    • Google Scholar
18. 1. Humphries JD
    2. Paul NR
    3. Humphries MJ
    4. Morgan MR

    (2015) Emerging properties of adhesion complexes: what are they and what do they do?

    Trends in Cell Biology 25:388–397.

    https://doi.org/10.1016/j.tcb.2015.02.008

    • Google Scholar
19. 1. Hynes RO

    (2002) Integrins

    Cell 110:673–687.

    https://doi.org/10.1016/S0092-8674(02)00971-6

    • Google Scholar
20. 1. Keren K
    2. Pincus Z
    3. Allen GM
    4. Barnhart EL
    5. Marriott G
    6. Mogilner A
    7. Theriot JA

    (2008) Mechanism of shape determination in motile cells

    Nature 453:475–480.

    https://doi.org/10.1038/nature06952

    • Google Scholar
21. 1. Khorshidi MA
    2. Vanherberghen B
    3. Kowalewski JM
    4. Garrod KR
    5. Lindström S
    6. Andersson-Svahn H
    7. Brismar H
    8. Cahalan MD
    9. Önfelt B

    (2011) Analysis of transient migration behavior of natural killer cells imaged in situ and in vitro

    Integrative Biology 3:770–778.

    https://doi.org/10.1039/c1ib00007a

    • Google Scholar
22. 1. Kim D-H
    2. Wirtz D

    (2013) Focal adhesion size uniquely predicts cell migration

    The FASEB Journal 27:1351–1361.

    https://doi.org/10.1096/fj.12-220160

    • Google Scholar
23. 1. Kim T-H
    2. Monsefi N
    3. Song J-H
    4. von Kriegsheim A
    5. Vandamme D
    6. Pertz O
    7. Kholodenko BN
    8. Kolch W
    9. Cho K-H

    (2015) Network-based identification of feedback modules that control RhoA activity and cell migration

    Journal of Molecular Cell Biology 7:242–252.

    https://doi.org/10.1093/jmcb/mjv017

    • Google Scholar
24. 1. Kiss A
    2. Gong X
    3. Kowalewski JM
    4. Shafqat-Abbasi H
    5. Strömblad S
    6. Lock JG

    (2015) Non-monotonic cellular responses to heterogeneity in talin protein expression-level

    Integr. Biol. 7:1171–1185.

    https://doi.org/10.1039/C4IB00291A

    • Google Scholar
25. 1. Kowalewski JM
    2. Shafqat-Abbasi H
    3. Jafari-Mamaghani M
    4. Endrias Ganebo B
    5. Gong X
    6. Strömblad S
    7. Lock JG
    8. Lee JW

    (2015) Disentangling membrane dynamics and cell migration; differential influences of f-actin and cell-matrix adhesions

    PLOS ONE 10:e0135204.

    https://doi.org/10.1371/journal.pone.0135204

    • Google Scholar
26. 1. Ku C-J
    2. Wang Y
    3. Weiner OD
    4. Altschuler SJ
    5. Wu LF

    (2012) Network crosstalk dynamically changes during neutrophil polarization

    Cell 149:1073–1083.

    https://doi.org/10.1016/j.cell.2012.03.044

    • Google Scholar
27. 1. Le Dévédec SE
    2. Yan K
    3. de Bont H
    4. Ghotra V
    5. Truong H
    6. Danen EH
    7. Verbeek F
    8. van de Water B

    (2010) Systems microscopy approaches to understand cancer cell migration and metastasis

    Cellular and Molecular Life Sciences 67:3219–3240.

    https://doi.org/10.1007/s00018-010-0419-2

    • Google Scholar
28. 1. Lee K
    2. Elliott HL
    3. Oak Y
    4. Zee C-T
    5. Groisman A
    6. Tytell JD
    7. Danuser G

    (2015) Functional hierarchy of redundant actin assembly factors revealed by fine-grained registration of intrinsic image fluctuations

    Cell Systems 1:37–50.

    https://doi.org/10.1016/j.cels.2015.07.001

    • Google Scholar
29. 1. Lewis L
    2. Verna JM
    3. Levinstone D
    4. Sher S
    5. Marek L
    6. Bell E

    (1982)

    The relationship of fibroblast translocations to cell morphology and stress fibre density

    Journal of Cell Science 53:21–36.

    • Google Scholar
30. 1. Li Z
    2. Lock JG
    3. Olofsson H
    4. Kowalewski JM
    5. Teller S
    6. Liu Y
    7. Zhang H
    8. Stromblad S

    (2010) Integrin-mediated cell attachment induces a PAK4-dependent feedback loop regulating cell adhesion through modified integrin v 5 clustering and turnover

    Molecular Biology of the Cell 21:3317–3329.

    https://doi.org/10.1091/mbc.E10-03-0245

    • Google Scholar
31. 1. Liu Y-J
    2. Le Berre M
    3. Lautenschlaeger F
    4. Maiuri P
    5. Callan-Jones A
    6. Heuzé M
    7. Takaki T
    8. Voituriez R
    9. Piel M

    (2015) Confinement and low adhesion induce fast amoeboid migration of slow mesenchymal cells

    Cell 160:659–672.

    https://doi.org/10.1016/j.cell.2015.01.007

    • Google Scholar
32. 1. Lock JG
    2. Wehrle-Haller B
    3. Strömblad S

    (2008) Cell–matrix adhesion complexes: master control machinery of cell migration

    Seminars in Cancer Biology 18:65–76.

    https://doi.org/10.1016/j.semcancer.2007.10.001

    • Google Scholar
33. 1. Lock JG
    2. Strömblad S

    (2010) Systems microscopy: an emerging strategy for the life sciences

    Experimental Cell Research 316:1438–1444.

    https://doi.org/10.1016/j.yexcr.2010.04.001

    • Google Scholar
34. 1. Lock JG
    2. Mamaghani MJ
    3. Shafqat-Abbasi H
    4. Gong X
    5. Tyrcha J
    6. Strömblad S
    7. Lee JW

    (2014) Plasticity in the macromolecular-scale causal networks of cell migration

    PLoS ONE 9:e90593.

    https://doi.org/10.1371/journal.pone.0090593

    • Google Scholar
35. 1. Lämmermann T
    2. Bader BL
    3. Monkley SJ
    4. Worbs T
    5. Wedlich-Söldner R
    6. Hirsch K
    7. Keller M
    8. Förster R
    9. Critchley DR
    10. Fässler R
    11. Sixt M

    (2008) Rapid leukocyte migration by integrin-independent flowing and squeezing

    Nature 453:51–55.

    https://doi.org/10.1038/nature06887

    • Google Scholar
36. 1. Lämmermann T
    2. Sixt M

    (2009) Mechanical modes of ‘amoeboid’ cell migration

    Current Opinion in Cell Biology 21:636–644.

    https://doi.org/10.1016/j.ceb.2009.05.003

    • Google Scholar
37. 1. Maiuri P
    2. Rupprecht J-F
    3. Wieser S
    4. Ruprecht V
    5. Bénichou O
    6. Carpi N
    7. Coppey M
    8. De Beco S
    9. Gov N
    10. Heisenberg C-P
    11. Lage Crespo C
    12. Lautenschlaeger F
    13. Le Berre M
    14. Lennon-Dumenil A-M
    15. Raab M
    16. Thiam H-R
    17. Piel M
    18. Sixt M
    19. Voituriez R

    (2015) Actin flows mediate a universal coupling between cell speed and cell persistence

    Cell 161:374–386.

    https://doi.org/10.1016/j.cell.2015.01.056

    • Google Scholar
38. 1. Masuzzo P
    2. Martens L
    3.  2014 Cell Migration Workshop Participants

    (2015) An open data ecosystem for cell migration research

    Trends in Cell Biology 25:55–58.

    https://doi.org/10.1016/j.tcb.2014.11.005

    • Google Scholar
39. 1. Moser M
    2. Legate KR
    3. Zent R
    4. Fassler R

    (2009) The tail of integrins, talin, and kindlins

    Science 324:895–899.

    https://doi.org/10.1126/science.1163865

    • Google Scholar
40. 1. Neumann B
    2. Walter T
    3. Hériché J-K
    4. Bulkescher J
    5. Erfle H
    6. Conrad C
    7. Rogers P
    8. Poser I
    9. Held M
    10. Liebel U
    11. Cetin C
    12. Sieckmann F
    13. Pau G
    14. Kabbe R
    15. Wünsche A
    16. Satagopam V
    17. Schmitz MHA
    18. Chapuis C
    19. Gerlich DW
    20. Schneider R
    21. Eils R
    22. Huber W
    23. Peters J-M
    24. Hyman AA
    25. Durbin R
    26. Pepperkok R
    27. Ellenberg J

    (2010) Phenotypic profiling of the human genome by time-lapse microscopy reveals cell division genes

    Nature 464:721–727.

    https://doi.org/10.1038/nature08869

    • Google Scholar
41. 1. Parsons JT
    2. Horwitz AR
    3. Schwartz MA

    (2010) Cell adhesion: integrating cytoskeletal dynamics and cellular tension

    Nature Reviews Molecular Cell Biology 11:633–643.

    https://doi.org/10.1038/nrm2957

    • Google Scholar
42. 1. Petrie RJ
    2. Koo H
    3. Yamada KM

    (2014) Generation of compartmentalized pressure by a nuclear piston governs cell motility in a 3D matrix

    Science 345:1062–1065.

    https://doi.org/10.1126/science.1256965

    • Google Scholar
43. 1. Petrie RJ
    2. Yamada KM

    (2015) Fibroblasts lead the way: a unified view of 3D cell motility

    Trends in Cell Biology 25:666–674.

    https://doi.org/10.1016/j.tcb.2015.07.013

    • Google Scholar
44. 1. Raftopoulou M
    2. Hall A

    (2004) Cell migration: rho GTPases lead the way

    Developmental Biology 265:23–32.

    https://doi.org/10.1016/j.ydbio.2003.06.003

    • Google Scholar
45. 1. Ridley AJ

    (2003) Cell migration: integrating signals from front to back

    Science 302:1704–1709.

    https://doi.org/10.1126/science.1092053

    • Google Scholar
46. 1. Riento K
    2. Ridley AJ

    (2003) Rocks: multifunctional kinases in cell behaviour

    Nature Reviews Molecular Cell Biology 4:446–456.

    https://doi.org/10.1038/nrm1128

    • Google Scholar
47. 1. Ruprecht V
    2. Wieser S
    3. Callan-Jones A
    4. Smutny M
    5. Morita H
    6. Sako K
    7. Barone V
    8. Ritsch-Marte M
    9. Sixt M
    10. Voituriez R
    11. Heisenberg C-P

    (2015) Cortical contractility triggers a stochastic switch to fast amoeboid cell motility

    Cell 160:673–685.

    https://doi.org/10.1016/j.cell.2015.01.008

    • Google Scholar
48. 1. Sahai E

    (2005) Mechanisms of cancer cell invasion

    Current Opinion in Genetics & Development 15:87–96.

    https://doi.org/10.1016/j.gde.2004.12.002

    • Google Scholar
49. 1. Sahai E
    2. Garcia-Medina R
    3. Pouyssegur J
    4. Vial E

    (2007) Smurf1 regulates tumor cell plasticity and motility through degradation of RhoA leading to localized inhibition of contractility

    The Journal of Cell Biology 176:35–42.

    https://doi.org/10.1083/jcb.200605135

    • Google Scholar
50. 1. Sanz-Moreno V
    2. Gadea G
    3. Ahn J
    4. Paterson H
    5. Marra P
    6. Pinner S
    7. Sahai E
    8. Marshall CJ

    (2008) Rac activation and inactivation control plasticity of tumor cell movement

    Cell 135:510–523.

    https://doi.org/10.1016/j.cell.2008.09.043

    • Google Scholar
51. 1. Schmitz MHA
    2. Held M
    3. Janssens V
    4. Hutchins JRA
    5. Hudecz O
    6. Ivanova E
    7. Goris J
    8. Trinkle-Mulcahy L
    9. Lamond AI
    10. Poser I
    11. Hyman AA
    12. Mechtler K
    13. Peters J-M
    14. Gerlich DW

    (2010) Live-cell imaging RNAi screen identifies PP2A–B55α and importin-β1 as key mitotic exit regulators in human cells

    Nature Cell Biology 12:886–893.

    https://doi.org/10.1038/ncb2092

    • Google Scholar
52. 1. Schwartz MA

    (2010) Integrins and extracellular matrix in mechanotransduction

    Cold Spring Harbor Perspectives in Biology 2:a005066.

    https://doi.org/10.1101/cshperspect.a005066

    • Google Scholar
53. 1. Schwarz US
    2. Gardel ML

    (2012) United we stand - integrating the actin cytoskeleton and cell-matrix adhesions in cellular mechanotransduction

    Journal of Cell Science 125:3051–3060.

    https://doi.org/10.1242/jcs.093716

    • Google Scholar
54. 1. Small JV
    2. Resch GP

    (2005) The comings and goings of actin: coupling protrusion and retraction in cell motility

    Current Opinion in Cell Biology 17:517–523.

    https://doi.org/10.1016/j.ceb.2005.08.004

    • Google Scholar
55. 1. Smilenov L
    2. Forsberg E
    3. Zeligman I
    4. Sparrman M
    5. Johansson S

    (1992) Separation of fibronectin from a plasma gelatinase using immobilized metal affinity chromatography

    FEBS Letters 302:227–230.

    https://doi.org/10.1016/0014-5793(92)80447-O

    • Google Scholar
56. 1. Somlyo AV
    2. Phelps C
    3. Dipierro C
    4. Eto M
    5. Read P
    6. Barrett M
    7. Gibson JJ
    8. Burnitz MC
    9. Myers C
    10. Somlyo AP

    (2003) Rho kinase and matrix metalloproteinase inhibitors cooperate to inhibit angiogenesis and growth of human prostate cancer xenotransplants

    The FASEB Journal 17:223–234.

    https://doi.org/10.1096/fj.02-0655com

    • Google Scholar
57. 1. Starke J
    2. Wehrle-Haller B
    3. Friedl P

    (2014) Plasticity of the actin cytoskeleton in response to extracellular matrix nanostructure and dimensionality

    Biochemical Society Transactions 42:1356–1366.

    https://doi.org/10.1042/BST20140139

    • Google Scholar
58. 1. Tadokoro S

    (2003) Talin binding to integrin tails: a final common step in integrin activation

    Science 302:103–106.

    https://doi.org/10.1126/science.1086652

    • Google Scholar
59. 1. Theisen U
    2. Straube E
    3. Straube A

    (2012) Directional persistence of migrating cells requires Kif1C-mediated stabilization of trailing adhesions

    Developmental Cell 23:1153–1166.

    https://doi.org/10.1016/j.devcel.2012.11.005

    • Google Scholar
60. 1. Welch MD

    (2015) Cell migration, freshly squeezed

    Cell 160:581–582.

    https://doi.org/10.1016/j.cell.2015.01.053

    • Google Scholar
61. 1. Winograd-Katz SE
    2. Itzkovitz S
    3. Kam Z
    4. Geiger B

    (2009) Multiparametric analysis of focal adhesion formation by RNAi-mediated gene knockdown

    The Journal of Cell Biology 186:423–436.

    https://doi.org/10.1083/jcb.200901105

    • Google Scholar
62. 1. Wolf K

    (2003) Compensation mechanism in tumor cell migration: mesenchymal-amoeboid transition after blocking of pericellular proteolysis

    The Journal of Cell Biology 160:267–277.

    https://doi.org/10.1083/jcb.200209006

    • Google Scholar
63. 1. Yatohgo T
    2. Izumi M
    3. Kashiwagi H
    4. Hayashi M

    (1988) Novel purification of vitronectin from human plasma by heparin affinity chromatography

    Cell Structure and Function 13:281–292.

    https://doi.org/10.1247/csf.13.281

    • Google Scholar
64. 1. Yoshida K

    (2006) Dissection of amoeboid movement into two mechanically distinct modes

    Journal of Cell Science 119:3833–3844.

    https://doi.org/10.1242/jcs.03152

    • Google Scholar

Author details

1. Hamdah Shafqat-Abbasi

   Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden

   Contribution

   HSA, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

2. Jacob M Kowalewski

   Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden

   Contribution

   JMK, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

3. Alexa Kiss

   Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden

   Contribution

   AK, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

4. Xiaowei Gong

   Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden

   Contribution

   XG, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

5. Pablo Hernandez-Varas

   Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden

   Contribution

   PHV, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

6. Ulrich Berge

   Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden

   Contribution

   UB, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

7. Mehrdad Jafari-Mamaghani

   Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden

   Contribution

   MJM, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

8. John G Lock

   Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden

   Contribution

   JGL, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Contributed equally with

   Staffan Strömblad

   For correspondence

   john.lock@ki.se

   Competing interests

   The authors declare that no competing interests exist.

9. Staffan Strömblad

   Department of Biosciences and Nutrition, Karolinska Institutet, Huddinge, Sweden

   Contribution

   SS, Analysis and interpretation of data, Drafting or revising the article

   Contributed equally with

   John G Lock

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0002-1236-6339

• 2,930

  views

• 570

  downloads

• 45

  citations

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