Catch and release: how do kinetochores hook the right microtubules during mitosis? TIGS-1107; No. of Pages 10 Catch and release: how do kinetochores hook the right microtubules during mitosis? Krishna K. Sarangapani and Charles L. Asbury Department of Physiology & Biophysics, University of Washington, Seattle, WA, USA Review Sport fishermen keep tension on their lines to prevent hooked fish from releasing. A molecular version of this angler’s trick, operating at kinetochores, ensures accu- racy during mitosis: the mitotic spindle attaches ran- domly to chromosomes and then correctly bioriented attachments are stabilized due to the tension exerted on them by opposing microtubules. Incorrect attachments, which lack tension, are unstable and release quickly, allowing another chance for biorientation. Stabilization of molecular interactions by tension also occurs in other physiological contexts, such as cell adhesion, motility, hemostasis, and tissue morphogenesis. Here, we review models for the stabilization of kinetochore attachments with an eye toward emerging models for other force- activated systems. Although attention in the mitosis field has focused mainly on one kinase-based mecha- nism, multiple mechanisms may act together to stabilize properly bioriented kinetochores and some principles governing other tension-sensitive systems may also apply to kinetochores. ‘‘Esa! Esa! Shame upon on you! You are but the pike, Kenozha, You are not the fish I wanted You are not the King of Fishes!’’ – Henry Wadsworth Longfellow, TheSongofHiawatha Tension-dependent stabilization of kinetochore– microtubule attachments Mitosis research has been guided for over half a century by the idea that mechanical tension signals proper attach- ment of chromosomes to microtubules of the mitotic spin- dle and selectively stabilizes these attachments. Chromosomes are coupled to spindle microtubules via kinetochores, which are multiprotein complexes that form persistent attachments to growing and shortening 0168-9525/$ – see front matter � 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tig.2014.02.004 Corresponding author: Asbury, C.L. (casbury@u.washington.edu). Keywords: cell division; mitotic spindle; meiosis; chromosome; biorientation; Aurora B; tension; mechanobiology; mechanosensor; catch bond. microtubule tips, thereby harnessing the dynamics of the filaments to produce force and movement. Accurate mitosis requires all kinetochores to become properly ‘bioriented’, with replicated sister chromatids attached to opposite sides of the spindle (or with homologous chromosomes attached to opposite sides during the first meiotic division of gametogenesis). Dietz [1] was the first to recognize that chromosomes repeatedly reorient on the spindle, in a trial- and-error process that ceases only when proper biorienta- tion is achieved because only this arrangement is stable [2,3]. Dietz also suggested a possible cause for this differ- ential stability: mechanical tension. Bioriented chromo- somes come under tension and their sister kinetochores are stretched apart by opposing spindle forces, whereas incorrectly attached chromosomes are relaxed (Figure 1A). Direct evidence that tension indeed confers stability to chromosome–spindle attachments came from classic micromanipulation experiments using grasshopper sper- matocytes [3]. The idea has since become a central tenet of mitosis research. An attractive molecular explanation for how tension may stabilize bioriented attachments began to emerge when genetic studies uncovered a kinase, Aurora B, whose activity prevents errors in chromosome segrega- tion [4–7]. Aurora B phosphorylates key microtubule- binding elements within the kinetochore [5,8–11], reduc- ing their biochemical affinity for [9,12], and promoting their detachment from, microtubules [12–16]. If the de- tachment-promoting activity of Aurora B is directed selectively toward kinetochores lacking tension and sup- pressed at kinetochores bearing tension, then it could explain why only relaxed attachments are unstable in vivo. Experiments in a variety of cell types are consistent with this idea (see especially [17–19]), but do not yet provide final proof (see below). Nevertheless, the hypoth- esis that tension suppresses Aurora B-triggered detach- ment has become so popular among mitosis researchers that it is difficult to find skepticism about it in current literature. Meanwhile, numerous instances have been uncovered in other physiological contexts where mechanical tension stabilizes molecular interactions. In some cases, force acts via regulation of kinase enzymes [20,21], as proposed for kinetochores. Other cases involve specialized ‘catch bonds’ that are directly stabilized by force [22–28]. Force on a protein can also promote or inhibit its proteolytic Trends in Genetics xx (2014) 1–10 1 http://dx.doi.org/10.1016/j.tig.2014.02.004 mailto:casbury@u.washington.edu Low tension High tension Microtubule Kinetochores Phosphorylated by Aurora B kinase Kinetochores Separated from Aurora B kinase Directly stabilized by tension Intrinsically unstable Pole Sister chroma�ds Sister kinetochores Incorrect (syntelic) Correct (bioriented) High tension Low tension Detach A�ach (A) (B) (C) TRENDS in Genetics Figure 1. Tension-dependent error avoidance during mitosis. (A) The accuracy of mitosis depends on trial-and-error and selective stabilization of correctly ‘bioriented’ attachments (i.e., those with sister kinetochores attached to microtubules emanating from opposite sides of the mitotic spindle). Bioriented kinetochores come under tension due to forces exerted on them by opposing microtubules, which somehow stabilizes the attachments. Conversely, a lack of tension on incorrectly attached kinetochores fails to stabilize them, so they release quickly, giving another chance for biorientation. For simplicity, we focus here on one type of incorrect attachment, called ‘syntelic’, with both sister kinetochores bound exclusively to microtubules from a single pole. Another incorrect attachment, ‘merotelic’, occurs when a single sister binds microtubules from both poles [83]. Although merotelics are geometrically distinct from syntelics, they may or may not be corrected by similar mechanisms [84]. (B,C) Two models for how tension may stabilize bioriented attachments. (B) Spatial separation model where a pool of Aurora B kinase located in-between the sister kinetochores (i.e., at the inner centromere) selectively phosphorylates the kinetochores of relaxed chromosomes, weakening their grip on the microtubules and promoting their release. Tension on correctly bioriented chromosomes causes them to stretch, spatially separating their kinetochores from Aurora B and preventing kinase-triggered detachment. Other mechanisms for tension-dependent suppression of Aurora B are also possible (Figure 2F,G). (C) Catch bond-like model where tension acts directly and independently of phosphoregulation on the kinetochore–microtubule interface, causing it to adopt a more stable configuration (Figure 2H). Review Trends in Genetics xxx xxxx, Vol. xxx, No. x TIGS-1107; No. of Pages 10 cleavage [29,30], or enable its binding to another protein [32,33]. Although understanding of how tension-activated molecules contribute to cell morphogenesis is far from complete, it is clear that many different schemes have evolved for sensing and responding to mechanical force [34–36]. Here, we review current ideas about how tension sta- bilizes kinetochore–spindle attachments. Key experi- ments that form the basis of the popular kinase-based model are examined. We also consider alternative mecha- nisms suggested by work outside the mitosis field and by recent experiments where reconstituted kinetochore–mi- crotubule attachments were directly manipulated in vitro. We do not discuss how the ‘wait anaphase’ (checkpoint) signals generated by kinetochores may be suppressed by tension, a topic already covered by several excellent reviews [37–39]. 2 Evidence for tension-dependent suppression of Aurora B kinase Aurora B is widely conserved, even across evolutionarily distant eukaryotes [40,41], and is clearly important for promoting proper attachments between chromosomes and spindle microtubules. Mutating it [4,5], depleting it from dividing cells [6], or inhibiting its activity [6,42] causes severe chromosome missegregation, although the spindle remains fully capable of attaching and pulling on kinetochores [5–7,17]. Many pairs of sister kineto- chores in Aurora B-deficient cells fail to biorient and the cells accumulate erroneous attachments where both sisters are bound to microtubules emanating from the same spindle pole [5–7]. In normal cells, such aberrant configurations are short lived [1–3], but Aurora B defi- ciency makes them unusually stable [42]. Aurora B phosphorylates key microtubule-binding elements within Review Trends in Genetics xxx xxxx, Vol. xxx, No. x TIGS-1107; No. of Pages 10 kinetochores in vivo, including the widely conserved Ndc80 and Knl1 subcomplexes [11,43,44] and the yeast Dam1 subcomplex [8]. Phosphorylation [9,12] and phosphomi- metic mutations at Aurora B target sites [15,16,44] reduce the biochemical affinity of kinetochore subcomplexes for microtubules [9,12,44] and also accelerate detachment of subcomplexes [12,16] and larger kinetochore assemblies [15] from microtubules in vitro. Altogether, these data sug- gest that Aurora B promotes detachment of kinetochores from microtubules and that this activity is somehow restrict- ed in vivo to erroneously attached kinetochores. The idea that tension might regulate Aurora B arose from studies of yeast engineered to enter mitosis without prior replication of their DNA or without sister chromatid cohesion [7,45]. The resulting unpaired chromatids cannot biorient, so the spindle exerts little or no tension on them, and they associate with either of the two spindle poles randomly [7]. However, upon Aurora B inhibition, they associate almost exclusively with the ‘old’ pole, that is, the one leftover from the previous cell cycle (rather than the ‘new’ pole formed de novo during the most recent S phase). Aurora B activity is evidently needed to break their attach- ments to the old pole, which they inherit from G1. This Aurora B-dependent turnover is reminiscent of the classic observations in insect spermatocytes, where unpaired X chromosomes undergo rapid pole-to-pole movements [46] because they lack the tension normally required for stabi- lizing chromosome–spindle attachments in these cells [2,3]. The similarity suggests a unified explanation: Aurora B may promote kinetochore detachment universally (dur- ing insect, yeast, and perhaps all eukaryotic cell division), and its activity may be universally inhibited by tension. The evidence that tension suppresses the detachment- promoting activity of Aurora B, although entirely circum- stantial, is compelling. Formally, any property absent from correctly attached kinetochores but shared by unpaired and erroneously attached kinetochores could underlie their differential susceptibility to detachment by Aurora B. Whatever the key difference, it does not depend on pre- cisely how kinetochore pairs are linked: bioriented kineto- chore pairs are resistant to Aurora B-dependent turnover regardless of whether they are linked naturally, through replicated sisters bound by cohesion [7], or artificially, through a single ‘dicentric’ DNA molecule or through DNA entanglements (created by inhibiting topoisomerase) [17]. This adaptability again mirrors the situation in insect spermatocytes, where erroneous attachments can be arti- ficially stabilized either by applying tension with a micro- needle [3] or by arranging a pair of maloriented chromosomes such that they become mechanically inter- locked [2]. Incorrect attachments can also be artificially stabilized in Drosophila S2 cells by overexpressing NOD [19], a kinesin-10 motor that localizes on chromosomes. Chromosome-anchored NOD is normally thought to help align chromosomes at the spindle equator by pushing them away from the poles. NOD overexpression may elevate this polar ejection force and significantly increase tension at kinetochores, thereby suppressing Aurora B and prevent- ing erroneous attachments from releasing [19]. Given the similarity of these observations, it seems likely that ten- sion somehow suppresses Aurora B. Notably, however, definitive evidence showing that direct application of me- chanical tension is sufficient to suppress chromosome re- orientation has only been obtained in meiotic grasshopper spermatocytes. Whether reorientation in these particular cells depends on Aurora B has not, to our knowledge, been proven. More generally, it remains unproven whether direct application of mechanical tension is sufficient to inhibit any Aurora B-dependent activity. Is kinetochore phosphorylation sensitive to tension? Immunostaining with the ‘3F3’ antibody confirms that ten- sion, either from a micromanipulation needle or from nor- mal spindle forces, can inhibit kinetochore phosphorylation [47–49]. 3F3 antibody specifically detects phosphorylated kinetochore proteins [50], but probably not Aurora B sub- strates. (It recognizes spindle checkpoint proteins phosphor- ylated either by Plk1 or Mps1 [51–54].) Nevertheless, key ideas that could apply to Aurora B are illustrated by experi- ments where chromosomes from lysed cells are washed, directly manipulated, and then immunostained with 3F3. Relaxed kinetochores on the washed chromosomes are de- void of 3F3 phosphoepitope, but they can be rephosphory- lated by incubation with ATP if a phosphatase inhibitor is also present [49]. Applying tension with a microneedle prevents rephosphorylation. The rephosphorylation of re- laxed kinetochores by incubation with ATP shows that the chromosomes retain a complete phosphorylation system, including substrate and kinase. The requirement for a phosphatase inhibitor shows that a phosphatase is also retained and does not require tension for its activity. Rather, tension must prevent phosphorylation in this case by inhi- biting the kinase, deforming the substrate, or repositioning the substrate relative to the kinase [49]. To our knowledge, such direct tests of tension sensitivity have not been performed using phosphospecific antibodies against bona fide Aurora B substrates. However, antibo- dies that recognize phosphorylation of several Aurora B substrates, including Ndc80 and Knl1, reveal correlations that are mostly consistent with tension-dependent sup- pression in vivo [11,44]. High levels of phosphorylation on Ndc80 and Knl1 correlate with unaligned, relaxed chromo- somes. Low phosphorylation is seen at metaphase, when most kinetochore pairs are aligned, stretched apart, and probably bioriented. An exception is anaphase, when low tension correlates with low Ndc80 phosphorylation [11], presumably because Aurora B delocalizes from the chro- mosomes at this time (binding instead to microtubules in the spindle midzone, together with other members of the ‘chromosomal passenger complex’ [41]). More puzzlingly, in cells treated with nocodazole to depolymerize their microtubules, the level of phosphorylation on Ndc80 is low [11], whereas high levels are seen for Knl1 [44]. Ten- sion should be absent after microtubule depolymerization, so the low phosphorylation of Ndc80 is incompatible with a strict model in which Aurora B always phosphorylates all its targets at relaxed kinetochores. The spatial separation model for suppression of Aurora B activity In mitotic animal cells, Aurora B localizes prominently on inner-centromeric chromatin, midway between sister 3 Review Trends in Genetics xxx xxxx, Vol. xxx, No. x TIGS-1107; No. of Pages 10 kinetochores. This pattern is the basis for an appealing ‘spatial separation’ model explaining how Aurora B may selectively phosphorylate the kinetochores of relaxed chro- mosomes: sister kinetochores on relaxed chromosomes are close enough to the inner centromere to be phosphorylated by Aurora B located there (Figure 1B). Tension stretches the chromosomes and this deformation spatially separates the kinetochores from centromeric Aurora B, thereby inhi- biting their phosphorylation. Aspects of this model have been tested using a Förster resonance energy transfer (FRET)-based biosensor for Au- rora B [18,44]. When the sensor is targeted to kinetochores (by fusion with the kinetochore proteins, Mis12 or Ndc80), its behavior matches the predictions for native kinetochore substrates, reporting high phosphorylation on unaligned kinetochores or after drug treatments that relax the chro- mosomes (e.g., nocodazole or monastrol), and reporting low phosphorylation on kinetochores that are properly aligned and bioriented. If instead the sensor is targeted to inner centromeres (by fusion with the centromere-targeting do- main of CENP-B) then its phosphorylation remains high, even on stretched chromosomes [18]. This observation indicates that the inner-centromeric Aurora B is constitu- tively active and that proximity of a substrate to this active pool is sufficient to cause its phosphorylation. The data are also nicely consistent with the spatial separation model, but a key question remains: in the normal physiological situation, is the Aurora B at the inner centromere directly responsible for phosphorylation of relaxed kinetochores? Two recent observations suggest that the Aurora B di- rectly involved in error correction is distinct from the prom- inent pool at the inner centromere. First, an antibody that specifically recognizes the phosphorylated active form of Aurora B labels not only the inner-centromeric pool between sister kinetochores, but also the outer kinetochore [11]. The outer-kinetochore population of Aurora B diminishes as kinetochores become properly aligned and less phosphory- lated on Ndc80, whereas the inner-centromeric pool remains prominent. The correlation between Ndc80 phosphorylation levels and enrichment of outer-kinetochore Aurora B sug- gests that this population, rather than the inner-centromer- ic pool, is responsible for phosphorylating Ndc80. A second recent observation is that Aurora B supports normal cell growth [55] and accurate chromosome segregation [56] even when its targeting to inner centromeres is disrupted (by mutations that interfere with the binding of INCENP to the centromere-targeting factor, survivin). Whether inner-cen- tromere localization of Aurora B is completely abolished in this case remains uncertain. However, if localization at the inner centromere is truly dispensable, then tension-depen- dent stretching of inner-centromeric chromatin cannot be the basis for suppression of Aurora B, and a key assumption of the spatial separation model would be wrong. Clearly, even with impressive advancements in understanding of Aurora B, we lack a complete picture of how tension confers stability to kinetochore–microtubule attachments. Mechanically regulated molecular systems outside of mitosis The notion that mechanical force can regulate molecular interactions and cellular activities is pervasive in many 4 fields besides mitosis. In a growing number of cases, ten- sion-dependent molecular behaviors have been demon- strated by the application of force directly to purified proteins. What can those of us interested in chromosome segregation learn from studies of other micromechanical systems (and vice versa)? One lesson is that nature has produced a variety of molecular mechanisms for sensing mechanical force (Figure 2A–E). There are examples where phosphorylation is regulated by force, similar to the proposal for kineto- chores, but not necessarily via spatial separation of the kinase from its substrates. Adherent cells, for example, sense their mechanical environment through a series of biochemical events that includes tension-dependent phos- phorylation of the focal adhesion protein p130Cas [57]. Tension acts in this case by ‘substrate priming’: the relaxed p130Cas substrate domain is normally resistant to phos- phorylation, but becomes susceptible under load (Figure 2E) [20]. Another phosphoregulatory mechanism occurs in muscle cells, where a kinase domain within the giant elastic protein titin senses mechanical strain and initiates a cascade of downstream biochemical events to control transcription of adaptive genes [58]. Force is trans- mitted directly through titin kinase itself, causing its activation by pulling an autoinhibitory domain away from its active (ATP-binding) site (Figure 2D) [21]. More gener- ally, force on a molecule can expose a previously buried ‘cryptic’ binding site for another molecule. Thus, stretching of the focal adhesion protein talin activates its binding to vinculin (Figure 2A) [32], and stretching an F-actin net- work activates the binding of b-integrin to the actin cross- linking protein, filamin [33]. Of course, force can also disrupt binding sites. The stretching that activates binding of b-integrin to filamin also causes the simultaneous un- binding of another filamin-binding protein, FliGAP [33]. Tension-dependent stabilization outside of mitosis As in mitosis, the concept that force can stabilize load- bearing molecular structures is central to many other areas of biology and the effect has been demonstrated in various molecular systems. Tension-dependent stabilization proba- bly explains how fibrils of collagen, the most abundant structural protein in vertebrates, are preferentially oriented along directions of load transmission [59]. Tension stabilizes collagen fibrils indirectly, by rendering them resistant to cleavage by collagenolytic enzymes [30], possibly because it inhibits the partial unfolding of the triple-helical structure of collagen (Figure 2C) [60]. Tension can also cause stabili- zation more directly, via specialized molecular interactions called ‘catch bonds’ (Box 1) [34,35]. Catch bonds were first demonstrated in single molecule experiments involving selectins [22], adhesion molecules that support tethering and rolling of leukocytes on vascular endothelium during inflammation. Lifetimes of individual selectin–ligand bonds initially increase and then decrease with tension, giving rise to a biphasic lifetime versus force curve [22,24]. Around the same time, it was also discovered that adhesion of fimbriat- ed bacteria to host cells is enhanced by hydrodynamic force [23], in part because a protein at the tip of the fimbria, fimH, forms catch bonds with mannosylated glycoproteins on the host cell surface. P Talin Vinculin Cryp�c site, exposed Without force With force Ti�n kinase ATP P Without force With force p130Cas Src kinase P P P Substrate access, gained Autoinhibited Ac�vated fimH Strengthened Mannose Autoinhibited Collagen Protected Proteolysis P P P Without force With force Aurora B kinase Micro- tubule Ch ro m os om e Kinetochore Substrate deforma�on inhibits phosphoryla�on Phosphoryla�on weakens microtubule binding P Kinase deforma�on inhibits phosphoryla�on P Bonds added or intrinsically strengthened (F) (A) (B) (D) (E) (C) (G) (H) Phosphoryla�on weakens microtubule binding TRENDS in Genetics Figure 2. Gallery of mechanosensitive molecules and models for mechanosensation at kinetochores. (A) Force exerted on talin exposes a cryptic binding site for vinculin [32]. (B) Force strengthens the fimH–mannose bond by pulling away an autoinhibitory domain [61]. (C) Force on collagen protects it from proteolytic cleavage [30]. (D) Force activates titin kinase by pulling away an autoinhibitory domain [21]. (E) Force on p130Cas promotes its phosphorylation by Src kinase [20]. (F–H) Speculative models for how force-dependent deformations could strengthen a kinetochore–microtubule attachment. (F) Kinetochore tension could deform Aurora B substrates, preventing the kinase from weakening their grip on the microtubule. (G) If Aurora B at the kinetochore–microtubule interface bears mechanical load, then the kinase itself could be inhibited by load. (H) Tension can also stabilize the kinetochore–microtubule interface directly [66], perhaps by altering the conformation of the microtubule tip or the kinetochore microtubule-binding elements in a way that strengthens existing bonds or promotes formation of additional bonds. Review Trends in Genetics xxx xxxx, Vol. xxx, No. x TIGS-1107; No. of Pages 10 Multiple force-sensitive molecules are often combined A second lesson that mitosis aficionados can learn from studies of other mechanically regulated systems is that multiple tension-controlled molecules often participate together in the same cellular process. Thus, bacterial adhesion is enhanced not only by fimH catch bonds at fimbrial tips [61], but also by mechanically responsive fimA molecules, which form a helical polymer comprising most of the length of the fimbriae. The fimA polymer acts as a near-perfect shock absorber, uncoiling and recoiling dynamically to maintain the optimal force at the tip (i.e., the force where fimH–mannose bonds are longest lived) [62]. Likewise, the blood clotting potential of von Will- ebrand Factor is regulated not only by the catch/flex bonds it forms with platelet glycoproteins [31,63,94], but also by a proteolytic cleavage process that is enhanced by shear forces [29,95]. Mechanosensation by adherent cells via integrins is perhaps the most multi- faceted example currently known. A dizzying number of force-sensitive molecules participate, including (i) force- activated binding of integrin to actin via filamin [33]; (ii) force-stabilized binding of integrin to fibronectin [64]; (iii) force-activated self-assembly of fibronectin [65]; (iv) force- activated binding of vinculin to talin [32]; and, possibly, (v) force-dependent regulation of focal adhesion kinase [36]. The cooperation of so many mechanically regulated proteins in one pathway may seem surprising. However, the experimental tools for studying force-sensitive mole- cules are relatively new. As they become more wide- spread, coincidence of multiple force sensors in a single pathway may turn out to be the norm rather than an exception. 5 Box 1. A conflagration of catch bonds Bell [85] first theorized that force would accelerate the dissociation of receptor–ligand bonds by tilting their energy landscape and lowering the energetic barrier for dissociation. Such interactions are known as ‘slip bonds’, and their lifetime typically decreases exponentially with tension. Later, it was proposed that tension could also do the opposite (i.e., prolong bond lifetime) by triggering a conformational change that tightens the ligand-binding pocket [86]. This counter-intuitive behavior can be likened to a finger trap gag toy: the harder one pulls, the more stable the interaction becomes. Since the initial discovery that force stabilizes selectin– ligand and fimH–mannose bonds (Figure 2B and main text), catch bonds have been found in many other biological contexts. Two prominent examples are the binding of integrin to fibronectin [64], which supports cell adhesion to the extracellular matrix, and the binding of von Willebrand factor via its A1 domain to platelet glycoprotein Ib [63,94], which initiates blood clotting preferentially in areas of high flow. Catch bond-like behavior is also common in ATP-powered motor proteins. Many muscle and nonmuscle myosins attach to actin filaments more stably when force opposing their motion is applied [25–28]. Opposing force generally prolongs the attachment lifetime of myosin by slowing its release of ADP and thereby preventing it from binding ATP, which is normally required for detachment of myosin from actin. The result is a ‘latch’ effect that allows these myosins to consume less ATP while sustaining loads for long durations (e.g., while maintaining vascular tone). A particularly dramatic example is myosin1b, whose actin-attachment lifetime increases >75-fold in response to small opposing loads (<2pN) [28]. Similarly, opposing force applied to the kinesin motor domain favors its tight binding to microtubules [87], in this case by accelerating its release of ADP (which has the opposite effect on kinesin as on myosin). This tension-dependent stabilization prob- ably helps to coordinate the hand-over-hand stepping of the twin motor domains of kinesin over the microtubule lattice [88]. Some of the earliest demonstrations of force-stabilized motor-filament attachments [25] pre-date the initial discovery of catch bonds. Most have not been described using the term ‘catch bond’, but their similarity is obvious. Review Trends in Genetics xxx xxxx, Vol. xxx, No. x TIGS-1107; No. of Pages 10 Kinetochore-microtubule attachments display catch bond-like behavior Given the diversity of force-sensing mechanisms across biology, it is natural to ask whether selective stabilization of proper kinetochore–microtubule attachments during mitosis relies solely on tension-dependent regulation of Aurora B, as is generally assumed, or whether other tension-dependent effects are also involved. Until recently, addressing this question would have been prohibitively difficult due to the lack of suitable in vitro assays for applying precisely controlled forces to kinetochore–micro- tubule interactions. Reconstitution of kinetochore–micro- tubule coupling using recombinant subcomplexes [13,16] and native kinetochore particles isolated from budding yeast [15,66] has made direct tests possible for the first time. One way to study purified kinetochore particles in vitro is to link them to polystyrene beads, which serve as artifi- cial cargoes (mimicking the chromosomes) and as handles to apply force (Figure 3A). Using a servo-controlled laser trap, bead-linked kinetochore particles can be attached to the tips of individual microtubules grown from coverslip- anchored seeds. The particles track with growing and shortening filament tips even when tension is applied continuously with the trap, to mimic the physiological situation. In addition, similar to kinetochores in vivo, 6 the purified particles can maintain persistent, load-bear- ing tip attachments through periods of microtubule growth and shortening [15,66]. To measure the effect of tension on attachment lifetime, the laser trap can be programed to operate as a force ‘clamp’, applying a chosen level of force for the duration of each event. A large number of events is recorded, and mean attachment lifetimes are calculated at each force by dividing the total observation time by the number of detachments. Considering that the purified kinetochore particles lack detectable Aurora B kinase [66], one might expect the lifetime to decrease monotoni- cally with force (similar to a typical ‘slip bond’; Box 1). However, attachment lifetimes initially increase and then decrease with tension, lending a biphasic shape to the lifetime versus force curve (Figure 3C) [66]. These results indicate the existence of a direct, catch bond-like stabili- zation mechanism that may act in parallel with Aurora- based phosphoregulation to help ensure mitotic accuracy. Catch bonds are often described by a two-state kinetic scheme where the bond switches between a weak and a strong state and force favors adoption of the strong state [24,67,68]. As force increases from zero, the average bond lifetime initially grows because the bond spends an in- creasing fraction of time in the strong state. Eventually, a critical force is reached, above which the lifetime decreases because the strong state is overpowered. This same model can describe the behavior of kinetochore–microtubule attachments (Figure 3B). Microtubule tips switch between two states, growth and shortening. Kinetochore particles detach from growing tips more slowly than from shortening tips (Figure 3E) [15,66], so tip growth corresponds to a strongly attached state. Moreover, tension decreases the likelihood that a growing tip will begin shortening (an event called a ‘catastrophe’) and increases the likelihood that a shortening tip will resume growth (a ‘rescue’), thereby causing the kinetochore-tip attachments to spend more time in the strongly attached state [66]. The resulting lifetime versus force relations for both kinetochore-micro- tubule attachments and canonical catch bond systems are well described by mathematically equivalent functions. However, the two systems differ in at least one way: canonical catch bond systems typically switch rapidly be- tween their weakly and strongly bound states, and because the associated conformational changes are subtle, it is not usually possible to directly measure kinetic rates for switching or detachment specifically from the strongly or weakly bound states. Given that microtubule tips in vitro switch relatively infrequently between growth and short- ening, distinguishing between the two states is straight- forward, and the specific kinetic rates can be directly measured [66]. Alternative mechanisms for regulation of kinetochore– spindle attachment stability The catch bond-like behavior observed using reconstituted kinetochore–microtubule attachments arises because their attachment stability depends on microtubule tip dynamics. This dependence implies more generally that altering microtubule tip dynamics by any means could affect kinet- ochore attachment stability. The core microtubule-binding kinetochore elements, Ndc80 and Dam1, for example, 0.01 0.1 1 10 100 Sw itc h ra te (h –1 ) 12840 Force (pN) 0.01 0.1 1 10 100 Sw itc h ra te (h –1 ) 12840 Force (pN) (D) (G) 0.1 1 10 100 D et ac hm en t r at e (h –1 ) 12840 Force (pN) 0.1 1 10 100 D et ac hm en t r at e (h –1 ) 12840 Force (pN) (E) (H) 50 40 30 20 10 0 Li fe �m e (m in ) 121086420 Force (pN) 50 40 30 20 10 0 Li fe �m e (m in ) 121086420 Force (pN) (C) (F) (A) (B) A�ached, growing �p A�ached, shortening �p Detachedk1 k2 k3 k4 k1 k2 k3 k4 k1 k2 k4 k3 Δx Laser trap Tens ion Mic ro- Kinetochore par�cle tub ule Bead TRENDS in Genetics Figure 3. Catch bond-like behavior of kinetochore–microtubule attachments, and how it may be regulated. (A) Schematic of laser trap assay. A bead decorated with native kinetochore particles or recombinant kinetochore subcomplexes is attached to the dynamic tip of a coverslip-anchored microtubule. As the microtubule tip grows and shortens, the kinetochore-bead moves with it. During bead movement, the laser trap can be automatically steered to keep a fixed bead-trap distance (Dx), thereby maintaining a constant tensile force on the kinetochore–microtubule interface. (B) A kinetochore-attached microtubule tip can grow (assemble) or shorten (disassemble), with transitions between these states governed by the rates of catastrophe, k1, and rescue, k2. The kinetochore can detach from either state, with rates k3 and k4. Given that k4 is generally much faster than k3, the overall rate of detachment can be reduced by inhibiting catastrophe (k1) or promoting rescue (k2). (C) The mean lifetime of reconstituted kinetochore–microtubule attachments initially increases and then decreases with force in a catch bond-like manner (gray curve, adapted from [66]). In principle, the lifetime versus force relation could be tuned to selectively stabilize relaxed attachments (blue) or those bearing higher loads (red). (D,E) Force inhibits catastrophe (k1, solid gray line), promotes rescue (k2, dashed gray), accelerates detachment during growth (k3, solid gray), and slows detachment during shortening (k4, dashed gray). Simultaneously adjusting the rates of catastrophe and detachment during growth across all forces as shown (blue and red lines in D and E) would shift the lifetime versus force relation leftward and rightward (blue and red curves in C). (F–H) Selective phosphorylation of relaxed kinetochores by Aurora B could accelerate detachment [green curve in (H)] or induce catastrophe [orange curve in (G)] at kinetochores that bear low forces (e.g., <4pN). In either case, the net effect would be to sharpen the lifetime versus force curve, increasing its sensitivity to force in the low-force regime where tension prolongs attachment [green and orange curves in (F)]. Review Trends in Genetics xxx xxxx, Vol. xxx, No. x TIGS-1107; No. of Pages 10 affect tip dynamics in ways that promote microtubule growth [16], which in turn promotes attachment stability. Tip stabilization by Ndc80 and Dam1 is partially reversed by phosphomimetic mutations at Aurora B target sites on these subcomplexes, suggesting that Aurora B promotes kinetochore release not only directly, by accelerating de- tachment, but also indirectly, by destabilizing kinetochore- attached microtubule tips [15,16]. Other candidates for affecting attachment stability via this mechanism are microtubule regulators of the Kinesin-13 [69] and Kine- sin-8 families [70], and plus end-binding proteins, such as XMAP215 and EB1 [71]. In principle, dividing cells might exploit the interplay between tension, microtubule tip dynamics, and kineto- chore attachment stability in interesting ways. Simulta- neously promoting catastrophes while inhibiting detachment during assembly, for example, is predicted to shift the lifetime versus force curve rightward (Figure 3C, red curve), so that the optimum force (where attachment lifetime is longest) occurs at higher tension. Conversely, inhibiting catastrophes while promoting de- tachment during assembly would shift the curve leftward. If the rate changes were large enough, then the catch-bond effect would be abolished, and lifetime would decrease monotonically with force (Figure 3C, blue curve). We spec- ulate that such shifts in the lifetime versus force relation may be biologically important. A leftward shift could allow stable attachment of relaxed kinetochores, perhaps pro- moting the formation of initial attachments during spindle assembly. A subsequent rightward shift could selectively destabilize kinetochores that have failed to biorient. A cell could tune the lifetime versus force relation for all kine- tochores simultaneously, by globally modulating the dy- namics of all kinetochore-attached microtubules [72]. The relation for a particular kinetochore could be tuned by post- translational modifications or by local binding of cofactors that alter detachment rates or the dynamics of attached microtubules. 7 Box 2. How tension-dependent modulation of microtubule tip dynamics could help stabilize biorientation If both sister kinetochores attach to the tips of microtubules emanating from the same spindle pole (syntelic, Figure 1A, main text), then opposing spindle forces will not develop and tension on both tip attachments will be low. Both tip attachments will tend to remain in the disassembling state and, consequently, both will be weak (Figure 3E, main text, k4>>k3). Correctly bioriented sister kinetochores will come under tension. In cells such as yeast, where microtubule minus ends are anchored statically at the spindle poles (i.e., in cells without poleward flux), the growth of tips attached to one bioriented sister kinetochore must be balanced by shortening of those attached to the other sister. This balance is probably achieved through tension-dependent modulation of tip dynamics [89–91]. The attachment on the growing side will be dramatically stabilized (Figure 3E, main text, k3<80 proteins, arranged in at least eight subcomplexes that as- semble hierarchically into a large structure [73,74]. Why all this molecular complexity? One possibility is that it enables kinetochores to sense and respond in a sophisticated manner to a variety of mechanical and biochemical cues. We imagine two alternative mechanisms, besides spa- tial separation, by which mechanical tension could sup- press Aurora B phosphorylation of kinetochores (Figure 2F,G). First, tension could deform the substrates of Aurora B within the kinetochore to render them inac- cessible to phosphorylation (or to render them more sus- ceptible to dephosphorylation by phosphatases). Consistent with this view, kinetochores exhibit deforma- tions that correlate with their attachment state [75–78] and could affect Aurora B kinase substrate access [39]. Tension-dependent control of substrate access at kineto- chores would be similar to the protection of tension-bearing collagen molecules against proteolytic cleavage [30,59] and to the priming of tension-bearing p130Cas for phosphory- lation by Src kinase [20]. A second possibility is that tension on Aurora B itself, or on its activator INCENP, could directly inhibit its kinase activity (Figure 2G). Auro- ra B and INCENP bind not only to chromatin and kine- tochores (as discussed above), but also to microtubules [79,80], so they could bear some mechanical load at kineto- chore–microtubule attachment sites. Direct mechanical control of Aurora B kinase activity would be similar to the load-dependent activation of titin kinase [21]. Assuming that Aurora B is indeed regulated by tension, how might it work together with the intrinsic catch bond- like behavior of kinetochores? The intrinsic catch bond-like behavior by itself produces only a modest stabilization in vitro (at the optimum force the mean attachment lifetime is increased approximately threefold relative to zero force) and the degree to which it will help stabilize biorientation in vivo is uncertain (Box 2). However, if Aurora B phos- phorylation has a strong enough effect on the rates of kinetochore detachment and microtubule switching, and if its effects are sufficiently inhibited by tension, then the lifetime versus force relation could be substantially sharp- ened. (Two examples of such sharpening are depicted in Figure 3F–H.) Another intriguing possibility is that kine- tochores may include canonical catch bonds that are stabi- lized by force even without changes in microtubule switching. If canonical catch bonds exist between kineto- chores and microtubules, they could have a profound influ- ence on attachment stability (Figure 2H) [81] and also chromosome movement [82]. 8 Concluding remarks In many ways, our understanding of how kinetochores ‘catch and hold’ the correct microtubules but release erro- neous attachments remains in its infancy. Various lines of evidence suggest that erroneous attachments are selec- tively eliminated through tension-dependent control of the spatial separation between Aurora B kinase and its kinet- ochore substrates. Considering the molecular complexity of kinetochores, the importance of mechanical cues during mitosis, and the diversity of force sensors in other areas of biology, it would not be surprising to find numerous other tension-sensing mechanisms operating at kinetochores as well. Precise mechanical manipulation of reconstituted kinetochore–microtubule attachments should enable di- rect tests of the popular kinase-based model and facilitate the search for additional tension-sensing mechanisms. Acknowledgments We thank Sue Biggins, Jeffrey Ruberti, Jason Stumpff, Linda Wordeman, Erik Yusko, Cheng Zhu, and four anonymous reviewers for their critical reading and helpful comments on this manuscript. We apologize to all those authors whose work we could not cite due to space limitations. Our work is currently supported by National Institutes of Health Grant RO1GM079373 and Packard Fellowship 2006-30521. References 1 Dietz, R. (1958) [Multiple sex chromosomes in Ostracoda cypria, their evolution and division characteristics]. Chromosoma 9, 359–440 2 Nicklas, R.B. (1974) Chromosome segregation mechanisms. Genetics 78, 205–213 3 Nicklas, R.B. and Koch, C.A. (1969) Chromosome micromanipulation. 3. Spindle fiber tension and the reorientation of mal-oriented chromosomes. J. Cell Biol. 43, 40–50 4 Chan, C.S. and Botstein, D. (1993) Isolation and characterization of chromosome-gain and increase-in-ploidy mutants in yeast. Genetics 135, 677–691 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0005 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0005 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0010 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0010 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0015 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0015 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0015 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0020 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0020 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0020 Review Trends in Genetics xxx xxxx, Vol. xxx, No. x TIGS-1107; No. of Pages 10 5 Biggins, S. et al. (1999) The conserved protein kinase Ipl1 regulates microtubule binding to kinetochores in budding yeast. Genes Dev. 13, 532–544 6 Hauf, S. et al. (2003) The small molecule Hesperadin reveals a role for Aurora B in correcting kinetochore-microtubule attachment and in maintaining the spindle assembly checkpoint. J. Cell Biol. 161, 281– 294 7 Tanaka, T.U. et al. (2002) Evidence that the Ipl1-Sli15 (Aurora kinase- INCENP) complex promotes chromosome bi-orientation by altering kinetochore-spindle pole connections. Cell 108, 317–329 8 Cheeseman, I.M. et al. (2002) Phospho-regulation of kinetochore- microtubule attachments by the Aurora kinase Ipl1p. Cell 111, 163–172 9 Cheeseman, I.M. et al. (2006) The conserved KMN network constitutes the core microtubule-binding site of the kinetochore. Cell 127, 983–997 10 DeLuca, J.G. et al. (2006) Kinetochore microtubule dynamics and attachment stability are regulated by Hec1. Cell 127, 969–982 11 DeLuca, K.F. et al. (2011) Temporal changes in Hec1 phosphorylation control kinetochore-microtubule attachment stability during mitosis. J. Cell Sci. 124, 622–634 12 Gestaut, D.R. et al. (2008) Phosphoregulation and depolymerization- driven movement of the Dam1 complex do not require ring formation. Nat. Cell Biol. 10, 407–414 13 Tien, J.F. et al. (2010) Cooperation of the Dam1 and Ndc80 kinetochore complexes enhances microtubule coupling and is regulated by aurora B. J. Cell Biol. 189, 713–723 14 Alushin, G.M. et al. (2012) Multimodal microtubule binding by the Ndc80 kinetochore complex. Nat. Struct. Mol. Biol. 19, 1161–1167 15 Sarangapani, K.K. et al. (2013) Phosphoregulation promotes release of kinetochores from dynamic microtubules via multiple mechanisms. Proc Natl Acad Sci U S A 110, 7282–7287 16 Umbreit, N.T. et al. (2012) The Ndc80 kinetochore complex directly modulates microtubule dynamics. Proc. Natl. Acad. Sci. U.S.A. 109, 16113–16118 17 Dewar, H. et al. (2004) Tension between two kinetochores suffices for their bi-orientation on the mitotic spindle. Nature 428, 93–97 18 Liu, D. et al. (2009) Sensing chromosome bi-orientation by spatial separation of aurora B kinase from kinetochore substrates. Science 323, 1350–1353 19 Cane, S. et al. (2013) Elevated polar ejection forces stabilize kinetochore-microtubule attachments. J. Cell Biol. 200, 203–218 20 Sawada, Y. et al. (2006) Force sensing by mechanical extension of the Src family kinase substrate p130Cas. Cell 127, 1015–1026 21 Puchner, E.M. et al. (2008) Mechanoenzymatics of titin kinase. Proc. Natl. Acad. Sci. U.S.A. 105, 13385–13390 22 Marshall, B.T. et al. (2003) Direct observation of catch bonds involving cell-adhesion molecules. Nature 423, 190–193 23 Thomas, W.E. et al. (2002) Bacterial adhesion to target cells enhanced by shear force. Cell 109, 913–923 24 Sarangapani, K.K. et al. (2004) Low force decelerates L-selectin dissociation from P-selectin glycoprotein ligand-1 and endoglycan. J. Biol. Chem. 279, 2291–2298 25 Veigel, C. et al. (2002) The gated gait of the processive molecular motor, myosin V. Nat. Cell Biol. 4, 59–65 26 Veigel, C. et al. (2003) Load-dependent kinetics of force production by smooth muscle myosin measured with optical tweezers. Nat. Cell Biol. 5, 980–986 27 Guo, B. and Guilford, W.H. (2006) Mechanics of actomyosin bonds in different nucleotide states are tuned to muscle contraction. Proc. Natl. Acad. Sci. U.S.A. 103, 9844–9849 28 Laakso, J.M. et al. (2008) Myosin I can act as a molecular force sensor. Science 321, 133–136 29 Zhang, X. et al. (2009) Mechanoenzymatic cleavage of the ultralarge vascular protein von Willebrand factor. Science 324, 1330–1334 30 Camp, R.J. et al. (2011) Molecular mechanochemistry: low force switch slows enzymatic cleavage of human type I collagen monomer. J. Am. Chem. Soc. 133, 4073–4078 31 Kim, J. et al. (2010) A mechanically stabilized receptor-ligand flex-bond important in the vasculature. Nature 466, 992–995 32 del Rio, A. et al. (2009) Stretching single talin rod molecules activates vinculin binding. Science 323, 638–641 33 Ehrlicher, A.J. et al. (2011) Mechanical strain in actin networks regulates FilGAP and integrin binding to filamin A. Nature 478, 260–263 34 McEver, R.P. and Zhu, C. (2010) Rolling cell adhesion. Annu. Rev. Cell Dev. Biol. 26, 363–396 35 Thomas, W.E. et al. (2008) Biophysics of catch bonds. Annu. Rev. Biophys. 37, 399–416 36 Moore, S.W. et al. (2010) Stretchy proteins on stretchy substrates: the important elements of integrin-mediated rigidity sensing. Dev. Cell 19, 194–206 37 Pinsky, B.A. and Biggins, S. (2005) The spindle checkpoint: tension versus attachment. Trends Cell Biol. 15, 486–493 38 Nezi, L. and Musacchio, A. (2009) Sister chromatid tension and the spindle assembly checkpoint. Curr. Opin. Cell Biol. 21, 785–795 39 Maresca, T.J. and Salmon, E.D. (2010) Welcome to a new kind of tension: translating kinetochore mechanics into a wait-anaphase signal. J. Cell Sci. 123, 825–835 40 Akiyoshi, B. and Gull, K. (2013) Evolutionary cell biology of chromosome segregation: insights from trypanosomes. Open Biol. 3, 130023 41 Carmena, M. and Earnshaw, W.C. (2003) The cellular geography of aurora kinases. Nat. Rev. Mol. Cell Biol. 4, 842–854 42 Lampson, M.A. et al. (2004) Correcting improper chromosome-spindle attachments during cell division. Nat. Cell Biol. 6, 232–237 43 Akiyoshi, B. et al. (2009) Analysis of Ipl1-mediated phosphorylation of the Ndc80 kinetochore protein in Saccharomyces cerevisiae. Genetics 183, 1591–1595 44 Welburn, J.P. et al. (2010) Aurora B phosphorylates spatially distinct targets to differentially regulate the kinetochore-microtubule interface. Mol. Cell 38, 383–392 45 Biggins, S. and Murray, A.W. (2001) The budding yeast protein kinase Ipl1/Aurora allows the absence of tension to activate the spindle checkpoint. Genes Dev. 15, 3118–3129 46 Nicklas, R.B. (1961) Recurrent pole-to-pole movements of the sex chromosome during prometaphase I in Melanoplus differentialis spermatocytes. Chromosoma 12, 97–115 47 Nicklas, R.B. et al. (1995) Kinetochore chemistry is sensitive to tension and may link mitotic forces to a cell cycle checkpoint. J. Cell Biol. 130, 929–939 48 Li, X. and Nicklas, R.B. (1997) Tension-sensitive kinetochore phosphorylation and the chromosome distribution checkpoint in praying mantid spermatocytes. J. Cell Sci. 110, 537–545 49 Nicklas, R.B. et al. (1998) Tension-sensitive kinetochore phosphorylation in vitro. J Cell Sci 111, 3189–3196 50 Gorbsky, G.J. and Ricketts, W.A. (1993) Differential expression of a phosphoepitope at the kinetochores of moving chromosomes. J. Cell Biol. 122, 1311–1321 51 Daum, J.R. et al. (2000) The 3F3/2 anti-phosphoepitope antibody binds the mitotically phosphorylated anaphase-promoting complex/ cyclosome. Curr. Biol. 10, R850–R852 52 Ahonen, L.J. et al. (2005) Polo-like kinase 1 creates the tension- sensing 3F3/2 phosphoepitope and modulates the association of spindle-checkpoint proteins at kinetochores. Curr. Biol. 15, 1078– 1089 53 Wong, O.K. and Fang, G. (2007) Cdk1 phosphorylation of BubR1 controls spindle checkpoint arrest and Plk1-mediated formation of the 3F3/2 epitope. J. Cell Biol. 179, 611–617 54 Conde, C. et al. (2013) Drosophila Polo regulates the spindle assembly checkpoint through Mps1-dependent BubR1 phosphorylation. Embo J. 32, 1761–1777 55 Yue, Z. et al. (2008) Deconstructing Survivin: comprehensive genetic analysis of Survivin function by conditional knockout in a vertebrate cell line. J. Cell Biol. 183, 279–296 56 Campbell, C.S. and Desai, A. (2013) Tension sensing by Aurora B kinase is independent of survivin-based centromere localization. Nature 497, 118–121 57 Tamada, M. et al. (2004) Activation of a signaling cascade by cytoskeleton stretch. Dev. Cell 7, 709–718 58 Lange, S. et al. (2005) The kinase domain of titin controls muscle gene expression and protein turnover. Science 308, 1599–1603 59 Flynn, B.P. et al. (2013) Highly sensitive single-fibril erosion assay demonstrates mechanochemical switch in native collagen fibrils. Biomech. Model. Mechanobiol. 12, 291–300 60 Chang, S.W. et al. (2012) Molecular mechanism of force induced stabilization of collagen against enzymatic breakdown. Biomaterials 33, 3852–3859 9 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0025 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0025 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0025 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0030 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0030 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0030 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0030 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0035 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0035 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0035 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0040 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0040 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0045 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0045 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0050 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0050 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0055 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0055 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0055 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0060 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0060 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0060 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0065 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0065 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0065 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0070 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0070 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0075 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0075 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0075 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0080 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0080 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0080 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0085 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0085 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0090 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0090 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0090 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0095 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0095 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0100 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0100 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0105 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0105 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0110 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0110 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0115 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0115 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0120 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0120 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0120 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0125 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0125 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0130 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0130 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0130 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0135 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0135 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0135 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0140 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0140 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0145 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0145 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0150 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0150 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0150 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0155 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0155 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0160 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0160 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0165 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0165 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0165 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0170 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0170 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0175 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0175 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0180 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0180 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0180 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0185 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0185 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0190 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0190 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0195 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0195 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0195 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0200 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0200 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0200 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0205 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0205 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0210 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0210 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0215 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0215 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0215 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0220 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0220 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0220 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0225 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0225 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0225 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0230 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0230 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0230 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0235 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0235 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0235 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0240 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0240 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0240 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0245 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0245 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0250 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0250 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0250 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0255 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0255 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0255 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0260 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0260 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0260 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0260 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0265 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0265 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0265 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0270 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0270 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0270 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0275 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0275 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0275 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0280 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0280 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0280 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0285 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0285 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0290 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0290 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0295 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0295 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0295 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0300 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0300 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0300 Review Trends in Genetics xxx xxxx, Vol. xxx, No. x TIGS-1107; No. of Pages 10 61 Yakovenko, O. et al. (2008) FimH forms catch bonds that are enhanced by mechanical force due to allosteric regulation. J. Biol. Chem. 283, 11596–11605 62 Forero, M. et al. (2006) Uncoiling mechanics of Escherichia coli type I fimbriae are optimized for catch bonds. PLoS Biol. 4, e298 63 Yago, T. et al. (2008) Platelet glycoprotein Ibalpha forms catch bonds with human WT vWF but not with type 2B von Willebrand disease vWF. J. Clin. Invest. 118, 3195–3207 64 Kong, F. et al. (2009) Demonstration of catch bonds between an integrin and its ligand. J. Cell Biol. 185, 1275–1284 65 Zhong, C. et al. (1998) Rho-mediated contractility exposes a cryptic site in fibronectin and induces fibronectin matrix assembly. J. Cell Biol. 141, 539–551 66 Akiyoshi, B. et al. (2010) Tension directly stabilizes reconstituted kinetochore-microtubule attachments. Nature 468, 576–579 67 Sarangapani, K.K. et al. (2011) Regulation of catch bonds by rate of force application. J. Biol. Chem. 286, 32749–32761 68 Thomas, W.E. (2009) Mechanochemistry of receptor-ligand bonds. Curr. Opin. Struct. Biol. 19, 50–55 69 Moore, A. and Wordeman, L. (2004) The mechanism, function and regulation of depolymerizing kinesins during mitosis. Trends Cell Biol. 14, 537–546 70 Su, X. et al. (2012) Move in for the kill: motile microtubule regulators. Trends Cell Biol. 22, 567–575 71 Howard, J. and Hyman, A.A. (2007) Microtubule polymerases and depolymerases. Curr. Opin. Cell Biol. 19, 31–35 72 Kabeche, L. and Compton, D.A. (2013) Cyclin A regulates kinetochore microtubules to promote faithful chromosome segregation. Nature 502, 110–113 73 Biggins, S. (2013) The composition, functions, and regulation of the budding yeast kinetochore. Genetics 194, 817–846 74 Takeuchi, K. and Fukagawa, T. (2012) Molecular architecture of vertebrate kinetochores. Exp. Cell Res. 318, 1367–1374 75 Wan, X. et al. (2009) Protein architecture of the human kinetochore microtubule attachment site. Cell 137, 672–684 76 Maresca, T.J. and Salmon, E.D. (2009) Intrakinetochore stretch is associated with changes in kinetochore phosphorylation and spindle assembly checkpoint activity. J. Cell Biol. 184, 373–381 77 Dumont, S. et al. (2012) Deformations within moving kinetochores reveal different sites of active and passive force generation. Science 337, 355–358 78 Joglekar, A.P. et al. (2009) In vivo protein architecture of the eukaryotic kinetochore with nanometer scale accuracy. Curr. Biol. 19, 694– 699 10 79 Kang, J. et al. (2001) Functional cooperation of Dam1, Ipl1, and the inner centromere protein (INCENP)-related protein Sli15 during chromosome segregation. J. Cell Biol. 155, 763–774 80 Sandall, S. et al. (2006) A Bir1-Sli15 complex connects centromeres to microtubules and is required to sense kinetochore tension. Cell 127, 1179–1191 81 Cane, S. et al. (2013) Insights from an erroneous kinetochore- microtubule attachment state. Bioarchitecture 3, 69–76 82 Civelekoglu-Scholey, G. et al. (2013) Dynamic bonds and polar ejection force distribution explain kinetochore oscillations in PtK1 cells. J. Cell Biol. 201, 577–593 83 Cimini, D. et al. (2003) Merotelic kinetochore orientation occurs frequently during early mitosis in mammalian tissue cells and error correction is achieved by two different mechanisms. J. Cell Sci. 116, 4213–4225 84 Cimini, D. et al. (2006) Aurora kinase promotes turnover of kinetochore microtubules to reduce chromosome segregation errors. Curr. Biol. 16, 1711–1718 85 Bell, G.I. (1978) Models for the specific adhesion of cells to cells. Science 200, 618–627 86 Dembo, M. et al. (1988) The reaction-limited kinetics of membrane-to- surface adhesion and detachment. Proc. R. Soc. Lond. B 234, 55–83 87 Uemura, S. and Ishiwata, S. (2003) Loading direction regulates the affinity of ADP for kinesin. Nat. Struct. Biol. 10, 308–311 88 Asbury, C.L. (2005) Kinesin: world’s tiniest biped. Curr. Opin. Cell Biol. 17, 89–97 89 Skibbens, R.V. et al. (1993) Directional instability of kinetochore motility during chromosome congression and segregation in mitotic newt lung cells: a push-pull mechanism. J. Cell Biol. 122, 859–875 90 Gardner, M.K. et al. (2005) Tension-dependent regulation of microtubule dynamics at kinetochores can explain metaphase congression in yeast. Mol. Biol. Cell 16, 3764–3775 91 Franck, A.D. et al. (2007) Tension applied through the Dam1 complex promotes microtubule elongation providing a direct mechanism for length control in mitosis. Nat. Cell Biol. 9, 832–837 92 Kwok, B.H. and Kapoor, T.M. (2007) Microtubule flux: drivers wanted. Curr. Opin. Cell Biol. 19, 36–42 93 Mitchison, T. et al. (1986) Sites of microtubule assembly and disassembly in the mitotic spindle. Cell 45, 515–527 [94] Ju, L. et al. (2013) The N-terminal flanking region of the A1 domain regulates the force-dependent binding of von Willebrand factor to platelet glycoprotein Ib alpha. J. Biol. Chem. 288, 32289–32301 [95] Wu, T. et al. (2010) Force-induced cleavage of single VWF A1A2A3 tridomains by ADAMTS-13. Blood 115, 370–378 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0305 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0305 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0305 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0310 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0310 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0315 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0315 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0315 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0320 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0320 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0325 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0325 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0325 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0330 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0330 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0335 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0335 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0340 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0340 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0345 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0345 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0345 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0350 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0350 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0355 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0355 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0360 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0360 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0360 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0365 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0365 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0370 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0370 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0375 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0375 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0380 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0380 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0380 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0385 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0385 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0385 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0390 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0390 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0390 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0395 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0395 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0395 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0400 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0400 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0400 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0405 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0405 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0410 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0410 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0410 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0415 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0415 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0415 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0415 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0420 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0420 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0420 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0425 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0425 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0430 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0430 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0435 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0435 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0440 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0440 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0445 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0445 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0445 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0450 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0450 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0450 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0455 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0455 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0455 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0460 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0460 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0465 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0465 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0470 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0470 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0470 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0475 http://refhub.elsevier.com/S0168-9525(14)00027-4/sbref0475 Catch and release: how do kinetochores hook the right microtubules during mitosis? Tension-dependent stabilization of kinetochore-microtubule attachments Evidence for tension-dependent suppression of Aurora B kinase Is kinetochore phosphorylation sensitive to tension? The spatial separation model for suppression of Aurora B activity Mechanically regulated molecular systems outside of mitosis Tension-dependent stabilization outside of mitosis Multiple force-sensitive molecules are often combined Kinetochore-microtubule attachments display catch bond-like behavior Alternative mechanisms for regulation of kinetochore-spindle attachment stability Aurora B on the horizon: new concepts in the light of mechanobiology Concluding remarks Acknowledgments References