Human cells express 45 kinesins, microtubule motors that transport a variety of molecules and organelles within the cell. Many kinesins also modulate the tracks they move on by either bundling or sliding or regulating the dynamic assembly and disassembly of the microtubule polymer. In migrating cells, microtubules control the asymmetry between the front and rear of the cell by differentially regulating force generation processes and substrate adhesion. Many of these functions are mediated by kinesins, transporters as well as track modulators. In this review, we summarize the current knowledge on kinesin functions in cell migration.
The kinesin superfamily
The first kinesin to be discovered, kinesin-1, was isolated as the translocator molecule in the squid giant axon . Since then, a superfamily comprising more than 40 kinesins in 14 families has been identified, many of the members of which are microtubule-dependent translocators such as kinesin-1. All kinesins contain a motor domain with ATPase activity and a microtubule-binding interface. The ATP hydrolysis cycle is coupled to changes in the affinity for microtubules and allows the motor to perform work. The tails of kinesins are highly diverse and bind cargo proteins, adapters and regulatory proteins. The motor domain can either be positioned at the N-terminus of the molecule for plus-end-directed motors such as kinesin-1, at the C-terminus in minus-end-directed motors of the kinesin-14 family or in the middle of the molecule such as in kinesin-13s. Many kinesins have adopted additional functions beyond transporting cargo along microtubules (Figure 1). For example, the homotetrameric kinesin-5 walks on two microtubules at the same time to separate the spindle poles at the beginning of mitosis , whereas kinesin-13s are specialized microtubule depolymerases that use ATP hydrolysis to induce microtubule disassembly, but can no longer walk along microtubules . This multitude of functions in the transport of cellular cargo along microtubules and the organization and dynamics of the microtubule cytoskeleton makes kinesins central players in many essential biological processes such as cell division, morphogenesis and locomotion.
Cell migration underlies embryonic development, immune responses and wound healing. Cell migration is stimulated by physical signals, such as the loss of cell–cell contacts and by chemical extracellular signalling through chemokine gradients. Directional migration requires the establishment and maintenance of a distinct front and rear of the cell and can be thought of as a cycle of four processes: protrusion at the leading edge, the formation of new adhesions at the leading edge, the release of adhesions at the rear of the cell and retraction of the cell rear (Figure 2). Cell migration is largely driven by the polymerization of actin filaments at the cell cortex and the myosin-mediated contraction of actin bundles [4,5]. Adhesion is mediated by integrin-rich adhesion structures that link extracellular fibres to the actin cytoskeleton . Microtubules are emerging as important regulators of the cell migration machinery, controlling force generation, substrate adhesion and signalling pathways [7,8]. Many of these functions are mediated or modulated by kinesins (Figure 3) and will be discussed in detail in the following sections.
Simplified model of cell migration
Kinesins transport cargo to the leading and trailing edge of the cell
Organization of the microtubule network in migrating cells
Many cells depend on the microtubule cytoskeleton to maintain the polarity of the migrating cell and for the asymmetric regulation of force generation and adhesion at the front and rear of the cell. The microtubule cytoskeleton itself needs to be asymmetric to control polarity. Most notable are the position of the centrosome in front of the nucleus and the majority of microtubules pointing with their plus-ends towards the leading edge of the cell. The minus-end-directed microtubule motor dynein is implicated in the positioning of the centrosome by exerting pulling forces on microtubules reaching the cell edge. The front-bias is achieved partly by the nucleation of front-directed microtubules from the trans-Golgi network  and partly by the differential regulation of microtubule dynamics in the front and rear of the cell. This not only regulates the microtubule number, but also the accumulation of post-translational modifications in front-directed microtubules. A number of kinesins preferentially transport cargo along microtubules carrying certain modifications [10–12]. This mechanism adds further front-directed bias to the microtubule network in migrating cells, thereby facilitating the efficient delivery of vesicles and molecular cargoes to the leading edge. Kinesins not only transport these cargoes, but also contribute to the organization of the microtubule network itself. For example, the kinesin-13 KIF2C/MCAK (mitotic centromere-associated kinesin) reduces the lifetimes of microtubules in the rear of the cell , whereas the kinesin-4 KIF4 stabilizes microtubules at the leading edge . Both these kinesins, thus, contribute to the differences in dynamics between front- and rear-facing microtubules. The kinesin-2 KIF3A regulates the orientation and dynamicity of front-directed microtubules . The inhibition of either of these kinesins results in decreased cell migration, which has been attributed to mis-regulation of microtubule organization and dynamics. Indeed, the suppression of dynamic growth and shrinkage excursions of microtubule ends using low doses of microtubule targeting agents impairs cell migration [16,17]. Thus kinesins are important regulators of microtubule dynamics whose activities serve to maintain a front-biased microtubule network.
Contribution to force-generation processes
As mentioned above, the actin cytoskeleton is usually credited with providing the driving forces for cell protrusion and contractility. However, microtubules are approximately 200 times stiffer than actin filaments  and, when laterally reinforced, can bear very high compressive loads . Application of microtubule depolymerizers to polarized epithelial cells or fibroblasts leads to the contraction of cells , suggesting that microtubules are important as a load-bearing element in cells. Furthermore, the sliding of microtubules in bundles by kinesin-1 motors generates pushing forces that drive the protrusion of neurites early in neuronal differentiation . Likewise, kinesin-5 KIF11/Eg5 has been implicated in cross-linking microtubules to prevent their penetration into parts of the neuronal growth cone. This process is important to relay the signal of path-finding cues to changing the direction of growth cone advance .
More indirectly, kinesins transport mRNAs encoding actin and components of the actin assembly-promoting machinery, as well as actin regulators themselves, to the leading edge. The accumulation of mRNA for actin, profilin and actin-related protein (Arp)2/3, the main nucleator of branched actin-filaments in lamellipodia at the leading edge, requires microtubule-based transport [23–25]. Localized translation of these mRNAs is required for efficient directionally-persistent cell migration as it determines the sites of actin nucleation . The kinesin-2 family members KIF3A and KIF17 bind to adenomatosis polyposis coli (APC) [26,27]. APC accumulates at the leading edge of the cell and stabilizes microtubules, but is also a potent actin nucleator itself and acts in synergy with formins to form long, unbranched actin filaments . The inhibition of kinesin-1 results in reduced lamellipodial protrusion . One potential cargo of kinesin-1 mediating this function is WAVE2, which depends on KIF5B for its localization to the leading edge of migrating MDA-MB-231 cells . WAVE2 promotes actin nucleation and branching by Arp2/3, with both processes being required for lamellipodia formation and protrusion . The identification of a kinesin-1 interaction motif has led to the discovery of more than 400 potential cargoes of kinesin-1 . It is, therefore, likely that kinesin-1 mediated transport stimulates cell protrusion through the delivery of a number of factors to the leading edge.
Regulation of cell adhesion
The importance of microtubules and kinesins in the regulation of cellular adhesion to the extracellular substrate has been known for over a decade. Microtubule ends repeatedly target focal adhesion sites and cause their dissolution . The depolymerization of microtubules or the inhibition of kinesin-1 leads to an increase in focal adhesion size [34,35]. Although clathrin-mediated endocytosis has been implicated in mediating the microtubule-dependent destabilization of focal adhesions, it is still unclear which cargoes are delivered by which kinesin to mediate these functions [36,37]. One possible mechanism is the release of adhesions from the extracellular matrix using secreted proteases. KIF5B and KIF3B transport the matrix metallopeptidase 9 (MMP-9) to the cell periphery where it is released to the extracellular environment and subsequently activated to allow the degradation of collagen and elastins, an activity essential for macrophage migration during the inflammatory response . KIF5B and KIF3A/KIF3B are involved in the surface exposure of the transmembranous MT1 (membrane type 1)-MMP , a collagenase responsible for the activation of other MMPs  as well as the degradation of the matrix components in close contact with the cell. Using in situ zymography, the kinesin-9 KIF9 has also been shown to be involved in matrix degradation , but which protease KIF9 transports remains to be identified.
Other kinesins promote cellular adhesion by contributing to the recycling of integrins, the major receptors of extracellular matrix proteins. The kinesin-3 KIF1C is involved in both focal adhesion maturation and formation of protruding adhesion structures called podosomes [42–44]. KIF1C participates in the maturation of trailing focal adhesions through its α5β1-integrin transport activity . α5β1-integrin is the main fibronectin receptor . When KIF1C transport is impaired, migrating retinal pigment epithelium cells fail to form mature focal adhesions at the rear, reducing their ability to maintain the directionality of migration . Recently, the kinesins KIF4A and KIF15 have also been implicated in integrin recycling: KIF15, a kinesin-12, is required for the internalization of α2β1-integrin, one of the most important collagen receptors , whereas kinesin-4 KIF4A transports α5β1-integrin into developing axons . Integrin recycling involves routes through different compartments and complex sorting mechanisms to fine-tune the set of displayed cell-surface receptors . It is very likely that kinesins play an important part in cargo sorting and directional transport for spatial exocytosis with the potential to differentially regulate substrate adhesion in different regions of the cell.
Besides integrins, other adhesion molecules are transported along microtubules. The kinesin-3 KIF14 transports adhesion proteins such as cadherin 11 (CDH11) and melanoma cell adhesion molecule (MCAM) to the cell surface, thereby decreasing the ability of cells to migrate and invade tissues . Consistent with this, KIF14 is a prognostic marker for the metastatic potential of cancer: lung adenocarcinoma expressing high levels of KIF14 develop fewer metastases than those expressing low levels of KIF14 . KIF14 over-expression also impairs integrin activation, leading to mis-regulated cell adhesion and cell migration .
Furthermore, kinesins have been implicated in the control of specialized matrix-remodelling adhesion structures called podosomes or invadosomes. Podosomes are protruding adhesion structures formed at the ventral side of some migratory and invasive cells that allows their trans-migration of the basement membrane. KIF1C and KIF9 are required for the formation as well as the turn-over of podosomes [41–43]. Depletion of either of these kinesins leads to a dramatic reduction in the number of podosomes formed by cells. KIF1C and KIF9 have been observed to decorate the plus-ends of a subset of microtubules targeting podosomes. When podosomes are contacted by microtubule ends, they adopt a dynamic behaviour, which results in the fusion or the fission of targeted podosomes .
Signalling pathways regulating cell migration
Recent studies highlighted the involvement of kinesins in the control of signalling pathways regulating cell migration. KIF3A over-expression in prostate cancer cells induces activation of the Wnt/β-catenin pathway  leading to an increase in cell proliferation as well as cell migration ability of these cells. This function could be mediated by APC, which controls the activity of β-catenin and binds to KIF3A .
Signalling is also promoted by kinesin-mediated receptor trafficking. In vascular smooth muscle cells, in response to vascular endothelial growth factor (VEGF) stimulation, the kinesin-3 KIF13B transports vesicles containing newly synthesized VEGF receptor 2 (VEGFR2) from the Golgi to the plasma membrane. The exposure of the receptor at the endothelial cell surface is required for VEGF-induced endothelial cell migration .
The kinesin-3 KIF14 negatively regulates integrin inside-out signalling . KIF14 interacts with Radil, an effector of Rap1. Rap1 is a small GTPase that mediates integrin activation and clustering in its activated GTP-bound form . KIF14 recruits Radil to microtubules and, thereby, controls the amount of Radil available for binding to Rap1-GTP at the plasma membrane . Thus KIF14 fine-tunes the balance between cell adhesion and cell migration.
Kinesins also bind to a number of signalling modules, scaffold proteins with important roles in signalling processes, which also mediate activation and cargo-loading of kinesins . It will be interesting to see to which extent these signalling proteins affect the activity of kinesin cargoes and modulate the signalling state at microtubule ends where cargo and motors tend to accumulate.
Kinesins are involved in most steps of cell migration, from force-generating processes that allow protrusion, to regulating adhesion and matrix degrading capabilities and the modulation of signalling pathways controlling migration. Until now, studies have tended to focus on one kinesin's involvement in one aspect of cell migration. How several kinesins work together and how the different cargoes of each kinesin contribute to its function in regulating cell migration is a key challenge for the future. Important requirements to successfully address this challenge will be the identification of the cargoes being transported by each kinesin and the microtubule sub-populations that serve as preferential tracks for their transport. Recent developments, including assays that facilitate kinesin-cargo identification  and attempts to crack the tubulin code [12,56], promise significant progress in this area in the coming years.
We thank Rob Cross for critical reading of the paper.
A.S. is a Lister Institute Research Prize Fellow. This work was supported by a British Heart Foundation non-clinical Ph.D. studentship [grant number FS/13/42/30377].
The Dynamic Cell: held at Robinson College, Cambridge University, Cambridge, U.K., 4–7 September 2014.