Cytoskeleton, molecular motors, and cell motility

Intuition Beginner

A cell is not a bag of dissolved chemicals. It has an internal scaffold — the cytoskeleton — built from three families of protein filaments. Threadlike actin filaments sit just under the membrane and let the cell crawl. Hollow microtubule tubes radiate outward from the cell's centre like a system of highways. Tough intermediate filaments brace the cell against tearing. Together they give the cell its shape, organise its contents, and power its movement.

Bolted to these filaments are molecular motors — protein machines that burn ATP fuel to walk. Myosin strides along actin. Kinesin walks toward the growing tip of a microtubule, and dynein walks the opposite way. A single kinesin takes steps of 8 nm, one step per ATP molecule, and can keep going for a micrometre before letting go. These motors deliver vesicles, drag chromosomes apart during division, and squeeze the cell in two.

The same machinery lets a cell migrate. To crawl, a cell pushes out a thin sheet of actin at its front edge, grips the surface through sticky junctions, and then hauls its body forward while releasing its grip at the back. Outside the cell, a mesh of proteins called the extracellular matrix provides the surface it crawls on, and junctions stitch neighbouring cells into tissues.

Visual Beginner

The three filament systems differ in size, in the motor proteins that use them as tracks, and in how dynamic they are. Actin and microtubules are polar — they carry a plus end and a minus end — and they grow and shrink. Intermediate filaments are rope-like and comparatively stable. The table collects the numbers that make the rest of the unit concrete.

Filament	Subunit	Diameter	Polar?	Associated motors	Typical role
Actin (microfilament)	G-actin monomer	~7 nm	Yes (+/−)	Myosin	Cortex, protrusion, contractile ring
Microtubule	αβ-tubulin heterodimer	~25 nm	Yes (+/−)	Kinesin (to +), Dynein (to −)	Spindle, intracellular tracks
Intermediate filament	Coiled-coil proteins (keratin, vimentin, lamin)	~10 nm	No	None known	Mechanical reinforcement

Worked example Beginner

Three numbers anchor the cytoskeleton's energy budget: the kinesin step size, the force a motor can push with, and how fast a filament grows.

Step 1 — a kinesin step. A kinesin-1 motor takes 8-nm steps along a microtubule, spending one ATP per step. At top speed it manages about 100 steps per second, so it travels $8\times 100=800\text{nm/s}$. To cross a 20-$\mu$m cell ($20000\text{nm}$) takes $20000/800=25$ seconds. One motor, one vesicle, a half-minute commute.

Step 2 — the force budget. A single kinesin or myosin motor stalls at about 5–7 pN of force. The mechanical work in one 8-nm step at a midpoint force of 6 pN is $W=F\times d=6\text{pN}\times 8\text{nm}=48\text{pN}\cdot \text{nm}$. A cell's ATP releases roughly $80$–$100\text{pN}\cdot \text{nm}$ of usable energy, so the motor converts about half of each ATP into directed work — a high efficiency for a nanometre-scale machine.

Step 3 — building a track. A microtubule grows by adding tubulin dimers at its plus end at roughly $1\mu \text{m/min}$, which is about $17\text{nm/s}$. Because each dimer adds 8 nm of length, that is about 2 dimers added per second. The cell keeps its actin monomer at a remarkably high concentration of about $500\mu \text{M}$ — roughly $3\times 10^5$ actin molecules per cubic micrometre — which is what makes rapid actin growth possible at the front of a crawling cell.

What this tells us: the cytoskeleton is a chemically powered mechanical system whose numbers — steps in nanometres, forces in piconewtons, concentrations in micromolar — are matched to the viscous, crowded interior of a living cell.

Check your understanding Beginner

Formal definition Intermediate+

A cytoskeletal filament is a linear protein polymer assembled from globular subunits. Two of the three systems — actin filaments and microtubules — are polar: their subunits sit in a head-to-tail arrangement, so the two ends are structurally distinct. By convention the faster-growing end is the plus end and the slower-growing end is the minus end. Polarity is what makes directed motor transport possible, because a motor head binds its track in a single orientation and steps toward one end only.

Monomer addition kinetics. Let $[M]$ be the free monomer concentration. For a given filament end with association rate constant $k_{+}$ (units ${\text{M}}^{-1}{\text{s}}^{-1}$) and dissociation rate $k_{-}$ (units ${\text{s}}^{-1}$), the net assembly flux is

$$J=k_{+}[M]-k_{-}.$$

The end grows when $J>0$ and shrinks when $J<0$. The critical concentration at that end is the monomer concentration at which growth and loss balance:

$$C_c=\frac{k_{-}}{k_{+}}.$$

For a polar filament, the two ends have different critical concentrations, $C_c^{+}$ and $C_c^{-}$. Treadmilling — net addition at the plus end balanced by net loss at the minus end, with no change in total filament length — occurs when $C_c^{+}<[M]<C_c^{-}$. Actin treadmills because ATP-bound actin adds readily at the plus end and, after hydrolysis to ADP-actin, dissociates preferentially from the minus end.

Dynamic instability of microtubules. Microtubules do not treadmill; they alternate stochastically between slow growth and rapid shrinkage. A growing microtubule carries a cap of GTP-bound tubulin at its plus end that stabilises the lattice. If the cap is lost (by hydrolysis outrunning addition), the filament switches to rapid shrinkage — a catastrophe. A shrinking filament can regain its GTP cap and resume growth — a rescue. The four parameters are the growth velocity $v_g$, the shrinkage velocity $v_s$ (typically several times larger), the catastrophe frequency $f_{\text{cat}}$, and the rescue frequency $f_{\text{res}}$.

Molecular motors are ATPases that couple the chemical cycle of ATP binding, hydrolysis, and product release to a mechanical displacement along a filament. The three families are myosin (walks on actin, with the myosin-II class driving muscle and the cytokinetic ring), kinesin (walks on microtubules, almost always toward the plus end), and dynein (walks on microtubules toward the minus end; the large axonemal dyneins also drive ciliary and flagellar beating). A motor's step size $d$ is the displacement per ATPase cycle (8 nm for kinesin-1 along a microtubule protofilament), its stall force $F_s$ is the maximum opposing force it can sustain (5–7 pN for kinesin-1 and myosin-V), and its run length is the mean distance travelled before detaching. Processivity is the ability to take many steps without dissociating, achieved when a motor has two motor domains so that at least one is always bound.

Cell-cell and cell-matrix junctions. Cells in a tissue are fastened to one another and to the extracellular matrix (ECM) by specialised junctions. Tight junctions (claudins, occludin) seal adjacent epithelial cells and block leakage. Adherens junctions (cadherins linked through catenins to actin) couple neighbouring contractile cortices. Desmosomes (desmoglein/desmocollin cadherins linked to intermediate filaments) resist shear. Hemidesmosomes (integrins linked to intermediate filaments) anchor epithelial cells to the underlying basement membrane. Focal adhesions (integrins linked to actin) anchor motile cells to the ECM and transmit force bidirectionally. Gap junctions (connexins) form aqueous channels that pass small molecules and ions directly between cells.

The extracellular matrix is the insoluble meshwork outside cells: fibrillar collagens (tensile strength), fibronectin and laminin (adhesive ligands), and proteoglycans (hydration and growth-factor storage). Cells contact it through integrins, transmembrane heterodimers that bind specific peptide motifs (the RGD sequence in fibronectin) and, on the cytoplasmic face, nucleate large platforms (focal adhesions) that connect to the actin cytoskeleton and to signalling pathways.

Rho-family GTPases set the polarity of a crawling cell by controlling which part of the actin cytoskeleton assembles where. Cdc42 drives filopodia (thin spike-like protrusions) and establishes cell polarity. Rac1 drives lamellipodia (broad sheet-like protrusions) through the Arp2/3 branching nucleator. RhoA activates formin-driven actin cables and myosin-II contractility, generating stress fibres and rear contraction. The three act in a spatial gradient — Rac1/Cdc42 at the front, RhoA at the back — to choreograph directional migration 17.07.01.

Key mechanism Intermediate+

The central mechanism of the entire cytoskeleton is the chemomechanical ATPase cycle of a molecular motor, in which the free energy of ATP hydrolysis is ratcheted into a directed step along a polar filament. State it for the actin-myosin cross-bridge, the same logic kinesin and dynein use on microtubules.

Mechanism (the actomyosin cross-bridge cycle). The myosin motor domain cycles through six nucleotide states, alternating tight and weak actin-binding, and translates one ATP of free energy into a displacement of the motor relative to the actin filament.

1. Rigor state. The myosin head is tightly bound to actin with no nucleotide. This is the state of a muscle after death (rigor mortis).
2. ATP binding. ATP enters the nucleotide pocket; the resulting conformational change reduces actin affinity and the head detaches.
3. Hydrolysis and recocking. ATP is hydrolysed to ADP and inorganic phosphate (Pi), both still bound. The released energy cocks the head into a strained, pre-stroke conformation ($\sim$60° tilt) displacing it $\sim$5 nm forward along actin.
4. Weak rebinding. The cocked head (ADP·Pi) binds actin weakly at a new actin monomer further toward the plus end.
5. Pi release and power stroke. Pi release tightens actin binding and triggers the power stroke: the head swings back to its relaxed tilt, driving the motor forward by $\sim$5 nm (muscle myosin-II) and generating force against any load.
6. ADP release. ADP leaves, returning the head to the rigor state, ready for the next ATP.

The energy budget is the load-bearing point. ATP hydrolysis in a cell releases free energy $\Delta {\mu }_{\text{ATP}}\approx 80$–$100\text{pN}\cdot \text{nm}$ ($\approx 20k_{\mathrm{B}}T$). The mechanical work delivered in a step of size $d$ against force $F$ is $W=Fd$. For kinesin-1, $d=8\text{nm}$ and $F_s\approx 6$–$7\text{pN}$, giving $W\approx 48$–$56\text{pN}\cdot \text{nm}$, i.e. an efficiency $\eta =W/\Delta {\mu }_{\text{ATP}}\approx 0.5$–$0.6$. A single motor cannot extract more work than the ATP supplies: the second law sets the upper bound $F_s\le \Delta {\mu }_{\text{ATP}}/d\approx 10$–$12\text{pN}$, consistent with the measured stall force.

Processivity is engineered by gating two motor domains. Kinesin-1 has two heads connected by a coiled-coil stalk. Strain through the stalk keeps the biochemical cycles out of phase: when one head binds ATP and steps forward, the partner is forced to hold its nucleotide until the new leading head has docked. This two-heads-interlocked gating means both heads almost never detach simultaneously, and the motor walks hundreds of steps (run length $\sim 1\mu \text{m}$) before dissociating. Yildiz and colleagues demonstrated the underlying hand-over-hand stepping directly by localising a fluorescent tag on one head to better than the step size [ref: TODO_REF Yildiz2004].

Bridge. The cross-bridge cycle is the foundational reason that chemical energy in ATP becomes directed mechanical work at molecular scale: a single kinesin or myosin head converts the $\sim 20k_{\mathrm{B}}T$ of ATP hydrolysis into an 8-nm step or a 5–7-pN stroke, and this is exactly the energy budget that all three filament systems exploit. The mechanochemical logic builds toward the cell-migration machinery of the Advanced results, where ensembles of actin polymerisation and myosin contraction are coordinated by Rho-family GTPases, and the same ATPase cycle appears again in muscle contraction 17.03.02 and in the mitotic spindle of the cell cycle 17.08.01. The central insight is that directionality is built into the polar filament track and read out by the motor head; putting these together, the bridge is from the ATPase chemistry of 17.01.02 pending to the mesoscale mechanics of moving cells.

Exercises Intermediate+

Advanced results Master

The Beginner picture treats each filament as a static cable and each motor as a lone walker. The Master picture is that the cytoskeleton is a far-from-equilibrium polymer system whose behaviour is set by the statistics of subunit exchange, and that motility emerges only when actin dynamics, adhesion, and contractility are coupled across a whole cell.

Dynamic instability as a GTP-cap phase transition. Mitchison and Kirschner's 1984 demonstration that individual microtubules switch abruptly between growth and shrinkage [ref: TODO_REF MitchisonKirschner1984], directly visualised by Horio and Hotani [ref: TODO_REF HorioHotani1986], is best understood through the GTP-cap model. The microtubule lattice is a tube of (typically) 13 protofilaments, each a head-to-tail string of αβ-tubulin dimers. Tubulin binds GTP, and GTP-tubulin adds to the plus end. After incorporation, tubulin hydrolyses its GTP to GDP. GDP-tubulin in the lattice is curved and lateral contacts between GDP-tubulin subunits are weaker, so a lattice of GDP-tubulin is mechanically strained and frays. Only a small cap of unhydrolysed GTP-tubulin at the plus end holds the tube together. If addition slows so that hydrolysis catches up — the cap vanishes — the end undergoes catastrophe, shrinking at $v_s\approx 10$–$30\mu \text{m/min}$, far faster than growth. A stochastic encounter with a GTP-tubulin island can re-cap the end (rescue). The result is that a population of microtubules explores space: individual filaments probe outward, those that hit a cap-stabilising structure (a kinetochore, a cortical anchor) are selected, the rest retract. This search-and-capture is how the mitotic spindle finds chromosomes.

Actin network dynamics and protrusion. The leading edge of a crawling cell is pushed not by a single filament but by a dendritic actin network nucleated by the Arp2/3 complex, which binds the side of an existing filament and starts a new branch at a characteristic 70° angle. Arp2/3 is activated by nucleation-promoting factors such as the WASP/WAVE family downstream of Cdc42 and Rac1. The network grows against the membrane by adding ATP-actin at the barbed (plus) end; the older, ADP-actin interior is severed by cofilin and recycled, and profilin recharges monomers with ATP. The combination — fast plus-end growth, Arp2/3 branching, cofilin severing, profilin recycling — produces a steady-state treadmilling network (not a single filament) that exerts a protrusive pressure of $\sim 1\text{nN}/\mu {\text{m}}^2$ on the membrane. Formins, in contrast, nucleate unbranched filaments and processively accelerate plus-end elongation, building the parallel bundles of filopodia and the contractile cables of stress fibres. Actin-binding proteins thus compose a kinetic machine: the Arp2/3 system makes a pushing gel; formins make pulling cables; cofilin turns the gel over; myosin-II pulls on it.

Motor processivity and gating. Single-molecule work has refined the kinesin mechanism quantitatively. Each 8-nm step is a hand-over-hand event: the rear head detaches, moves past the docked front head by $16\text{nm}$ (twice the step size), and rebinds, while the centre of mass advances $8\text{nm}$ [ref: TODO_REF Yildiz2004]. Gating is mechanical: the two heads are linked by a neck linker, and docking of the neck linker on the leading head upon ATP binding throws the rear head forward; conversely, inter-head strain suppresses ATP binding on the trailing head until the leading head has rebind. The run length is $\sim 1\mu \text{m}$ (about 125 steps), corresponding to a per-step detachment probability of $\sim 0.008$. The force-velocity curve is approximately linear from $v_{\max }\approx 800\text{nm/s}$ at zero load down to zero velocity at stall $F_s\approx 6$–$7\text{pN}$, with backward stepping above stall. Dynein, by contrast, is a huge ring-shaped AAA+ ATPase whose step size is variable ($8$–$32\text{nm}$) and direction is regulated by the dynactin complex and cargo adaptors.

The cell-migration cycle. Abercrombie's three-step model — protrusion, attachment (grip), and traction with rear release — is now resolved molecularly. (1) Protrusion. Cdc42/Rac1 activation at the leading edge nucleates dendritic actin via Arp2/3, pushing the lamellipodium outward; nascent adhesions form under it. (2) Adhesion maturation. Integrins cluster, recruit talin, vinculin, and focal adhesion kinase (FAK), and link to actin stress fibres. As the cell moves forward, the adhesion matures into a focal adhesion that transmits traction; the force itself stiffens the adhesion (mechanosensitivity through vinculin unfolding and talin exposure of vinculin-binding sites). (3) Translocation and rear release. RhoA/myosin-II contraction in the cell body generates traction on mature adhesions, pulling the soma forward. At the rear, calpain- and FAK/Src-mediated disassembly releases adhesions. The cycle is polarised by a front-rear gradient of PIP3 (PIP kinase activated by PI3K at the front 17.07.03 pending), Rho GTPases, and the microtubule organising centre, which is repositioned toward the front and delivers cargo that stabilises the leading edge. Durotaxis, the tendency of cells to crawl toward stiffer substrate, falls out of this cycle: a cell grips a stiff matrix more strongly, exerts more traction, and is pulled toward the stiffness gradient. Collective migration — sheets and strands of cells — adds cell-cell cadherin junctions to the same machinery, so that a "leader" cell drags its followers.

Synthesis. Filament dynamics, motor mechanochemistry, and adhesion are the foundational reason a eukaryotic cell can both hold its shape and reshape itself: actin treadmilling and microtubule dynamic instability supply polar tracks and a search-and-capture system, and this is exactly the substrate on which myosin, kinesin, and dynein read out ATP. The motility cycle generalises from a single fibroblast on glass to wound closure, immune surveillance, neural-crest migration, and cancer invasion, and is dual to the cytokinetic-ring logic of cell division in that both couple protrusion to adhesion and myosin-driven contraction. The central insight is that cell movement is a spatially polarised chemical reaction cycle — polymerisation at the front, depolymerisation and contraction at the rear — choreographed by Rho GTPases and PIP lipids; putting these together, the bridge is from single-molecule mechanics to tissue-scale morphogenesis, and the bridge is reinforced by the same filaments and motors appearing again in muscle contraction 17.03.02 and in the mitotic spindle 17.08.01.

Full proof set Master

Proposition (treadmilling steady state). Let a polar filament have plus-end rate constants $(k_{+}^{+},k_{-}^{+})$ and minus-end constants $(k_{+}^{-},k_{-}^{-})$, with critical concentrations $C_c^{+}=k_{-}^{+}/k_{+}^{+}$ and $C_c^{-}=k_{-}^{-}/k_{+}^{-}$, and suppose $C_c^{+}<C_c^{-}$. At the free-monomer concentration

$$[M{]}_{\text{tread}}=\frac{k_{-}^{+}+k_{-}^{-}}{k_{+}^{+}+k_{+}^{-}},$$

the filament treadmills: it gains subunits at the plus end and loses subunits at the minus end at equal rates, so its total length is constant while subunits stream through the filament from plus to minus. The treadmilling flux is $J_{\text{tread}}=k_{+}^{+}[M{]}_{\text{tread}}-k_{-}^{+}$.

Proof. The plus end grows at rate $J_{+}=k_{+}^{+}[M]-k_{-}^{+}$ and the minus end changes length at rate $J_{-}=k_{+}^{-}[M]-k_{-}^{-}$ (positive meaning growth). The total filament length is constant precisely when the plus-end gain equals the minus-end loss, i.e. $J_{+}=-J_{-}$, or equivalently $J_{+}+J_{-}=0$:

$$(k_{+}^{+}+k_{-}^{-}\cdot 0)\to (k_{+}^{+}+k_{+}^{-})[M]-(k_{-}^{+}+k_{-}^{-})=0.$$

Rearranging gives the stated $[M{]}_{\text{tread}}=(k_{-}^{+}+k_{-}^{-})/(k_{+}^{+}+k_{+}^{-})$. At this concentration the plus-end gain is $J_{\text{tread}}=k_{+}^{+}[M{]}_{\text{tread}}-k_{-}^{+}=\frac{k_{+}^{+}(k_{-}^{+}+k_{-}^{-})}{k_{+}^{+}+k_{+}^{-}}-k_{-}^{+}=\frac{k_{+}^{+}{k}_{-}^{-}-k_{+}^{-}{k}_{-}^{+}}{k_{+}^{+}+k_{+}^{-}}$. Hence $J_{\text{tread}}>0$ requires $k_{+}^{+}{k}_{-}^{-}>k_{+}^{-}{k}_{-}^{+}$, i.e. $k_{+}^{+}/k_{-}^{+}>k_{+}^{-}/k_{-}^{-}$, equivalently $1/C_c^{+}>1/C_c^{-}$, i.e. $C_c^{+}<C_c^{-}$. Thus treadmilling at constant length is possible exactly in the regime asserted, and the flux is as stated. $\square$

Proposition (stall-force bound). A molecular motor that advances by a step of size $d$ while consuming exactly one ATP of free energy $\Delta {\mu }_{\text{ATP}}$ per step cannot sustain a stall force greater than $F_s^{\max }=\Delta {\mu }_{\text{ATP}}/d$. The thermodynamic efficiency at stall is $\eta =F_{s}d/\Delta {\mu }_{\text{ATP}}\le 1$.

Proof. Treat one ATPase cycle as a thermodynamic engine operating isothermally at temperature $T$. The free energy available from the cell's ATP pool in one cycle is $\Delta {\mu }_{\text{ATP}}>0$ (set by the ATP/ADP·Pi ratio, typically $\Delta {\mu }_{\text{ATP}}\approx 80$–$100\text{pN}\cdot \text{nm}$ in cytoplasm). The mechanical work the motor delivers against an opposing load $F$ in one step is $W=Fd$. The second law requires that the motor cannot, in a cyclic process, deliver more work than the free energy it consumes: $W\le \Delta {\mu }_{\text{ATP}}$. Hence $Fd\le \Delta {\mu }_{\text{ATP}}$, giving $F\le \Delta {\mu }_{\text{ATP}}/d$. At stall the motor's velocity is zero and it holds its position; the largest load it can hold without slipping backward satisfies $F_s\le \Delta {\mu }_{\text{ATP}}/d$. Defining efficiency as $\eta =W/\Delta {\mu }_{\text{ATP}}=F_{s}d/\Delta {\mu }_{\text{ATP}}$, the same inequality gives $\eta \le 1$. For kinesin-1, $d=8\text{nm}$ and $\Delta {\mu }_{\text{ATP}}\approx 90\text{pN}\cdot \text{nm}$ predict $F_s^{\max }\approx 11\text{pN}$; the measured stall force of $6$–$7\text{pN}$ corresponds to $\eta \approx 0.5$–$0.6$, well below the bound and consistent with the irreversibility of product release. $\square$

Proposition (run length from processivity). Suppose a motor steps forward by distance $d$ each cycle, and after each completed step detaches with probability $1-p$ (so it continues with probability $p$). The mean run length is

$$L=\frac{dp}{1-p}.$$

Proof. Let $L$ denote the mean distance travelled from a freshly attached state. With probability $1-p$ the motor detaches immediately after the first step, contributing distance $d$. With probability $p$ it completes a step (distance $d$) and then finds itself in a fresh attached state, from which it again travels a mean distance $L$. Hence $L=(1-p)d+p(d+L)=d+pL$. Solving, $L(1-p)=dp$, so $L=dp/(1-p)$. For kinesin-1 with $d=8\text{nm}$ and measured mean run length $L\approx 1\mu \text{m}=1000\text{nm}$, one recovers $p/(1-p)=125$, i.e. $p\approx 0.992$: the motor detaches on roughly one step in 125. $\square$

These three results fix the quantitative spine of the unit: the first the steady polymerisation flux that powers protrusion, the second the thermodynamic ceiling on what any single ATP can buy, the third the statistics that connect a microscopic detachment probability to the mesoscopic run length an optical trap measures.

Connections Master

• Cellular organization and organelles 17.03.01 is the structural prerequisite: microtubules radiate from the centrosome and serve as the tracks along which kinesin and dynein deliver ER- and Golgi-derived vesicles to their target compartments, and the secretory pathway described there is the cargo that this unit's motors actually haul.

• Cytoskeleton and contractile proteins 17.03.02 is the closest sibling: it treats actin-myosin cross-bridge mechanics and force-velocity in muscle, and hooks upward to organismal musculoskeletal biology. This unit complements it by emphasising filament dynamics, the three motor superfamilies, and the cell-junction, ECM, and cell-migration machinery that tissue-level contractility is built from.

• Protein structure and folding 17.01.02 pending supplies the molecular grammar on which motors run: the motor head is an ATPase enzyme whose nucleotide pocket closes around ATP and transmits a conformational change through a converter domain and neck linker to the actin- or tubulin-binding interface, exactly the structure-mechanism coupling catalogued there.

• Cell signalling — receptors and small GTPases 17.07.01 and PI3K/Akt/mTOR 17.07.03 pending set the regulatory context: Rho-family GTPases (Cdc42, Rac1, RhoA) polarise the actin cytoskeleton and define the front and rear of a migrating cell, while PIP3 produced by PI3K at the leading edge recruits the Rac1 machinery that nucleates the lamellipodium.

• Cell and molecular biology methods 17.11.01 provides the instruments that produced nearly every number in this unit: fluorescence microscopy and single-molecule tracking measured kinesin's 8-nm steps, optical tweezers measured the 5–7-pN stall force, and super-resolution localisation imaged the Arp2/3-branched actin network at the leading edge.

• The cell cycle 17.08.01 consumes this unit's machinery wholesale: microtubule dynamic instability builds and searches with the mitotic spindle, kinesin and dynein generate the forces that align and separate chromosomes, and an actin-myosin contractile ring executes cytokinesis.

• Cellular neuroscience 17.09.01 depends on microtubule-based axonal transport: kinesin carries synaptic-vesicle precursors and membrane proteins from the soma down the axon toward the synapse, and dynein returns damaged organelles and neurotrophic signals — failures of these motors underlie axonal transport diseases.

Historical & philosophical context Master

The cytoskeleton was discovered twice: first as the banded scaffolding of muscle, then as the motile machinery of every cell. Hugh Huxley's 1969 sliding-filament and cross-bridge model [ref: TODO_REF Huxley1969] synthesised a decade of X-ray diffraction and electron microscopy into the proposal that muscle contraction is the cyclic interaction of myosin cross-bridges pulling on actin filaments, with each stroke powered by ATP. Huxley's insight — that force generation and filament sliding are separate, ATP-coupled events — remains the conceptual core of all motor biology.

The generalisation from muscle to all cells took a further fifteen years and a methodological leap. Michael Sheetz and James Spudich's 1983 in vitro motility assay [ref: TODO_REF SheetzSpudich1983] coated fluorescent beads with myosin and watched them move along actin cables of the alga Nitella, proving that actin-myosin motility needs nothing more than the two proteins and ATP. Ron Vale, Thomas Reese, and Michael Sheetz's 1985 isolation of kinesin [ref: TODO_REF Vale1985] from squid giant axon — undertaken to explain fast axonal transport — identified a motor that walked on microtubules rather than actin, opening the kinesin and dynein superfamilies. Within a year, the premise that cells carry a fleet of ATP-fuelled walking proteins was established fact.

The second pillar — filament dynamics — was set by Tim Mitchison and Marc Kirschner's 1984 paper on dynamic instability [ref: TODO_REF MitchisonKirschner1984], which proposed that microtubules switch stochastically between growth and rapid shrinkage, against the then-prevailing view of slow equilibrium polymerisation. The prediction was confirmed directly by Hotani and Horio's dark-field microscopy of individual microtubules [ref: TODO_REF HorioHotani1986], which showed the abrupt transitions. The deeper lesson is that biological polymers are not passive cables but exploratory, far-from-equilibrium systems that the cell uses to search space — most strikingly in the mitotic spindle's capture of chromosomes.

The single-molecule era, from the 1990s optical-tweezers measurements of kinesin stall force to Paul Selvin and colleagues' 2004 demonstration that kinesin walks hand-over-hand [ref: TODO_REF Yildiz2004] by localising a fluorophore on one head to nanometre precision, closed the gap between the ensemble biochemistry of Huxley's era and the discrete mechanical events of a single motor. Cell motility invokes no new force law: it is the statistics and chemistry of polymers and ATPases, operating at the length and energy scale where thermal fluctuation and directed work are comparable.

Bibliography Master

@article{Huxley1969,
  author = {Huxley, H. E.},
  title  = {The mechanism of muscular contraction},
  journal= {Science},
  volume = {164},
  pages  = {1356--1366},
  year   = {1969}
}

@article{SheetzSpudich1983,
  author = {Sheetz, M. P. and Spudich, J. A.},
  title  = {Movement of myosin-coated fluorescent beads on actin cables in vitro},
  journal= {Nature},
  volume = {303},
  pages  = {31--35},
  year   = {1983}
}

@article{Vale1985,
  author = {Vale, R. D. and Reese, T. S. and Sheetz, M. P.},
  title  = {Identification of a novel force-generating protein, kinesin, involved in microtubule-based motility},
  journal= {Cell},
  volume = {42},
  pages  = {39--50},
  year   = {1985}
}

@article{MitchisonKirschner1984,
  author = {Mitchison, T. and Kirschner, M.},
  title  = {Dynamic instability of microtubule growth},
  journal= {Nature},
  volume = {312},
  pages  = {237--242},
  year   = {1984}
}

@article{HorioHotani1986,
  author = {Horio, T. and Hotani, H.},
  title  = {Visualization of the dynamic instability of individual microtubules by dark-field microscopy},
  journal= {Nature},
  volume = {321},
  pages  = {605--607},
  year   = {1986}
}

@article{Yildiz2004,
  author = {Yildiz, A. and Tomishige, M. and Vale, R. D. and Selvin, P. R.},
  title  = {Kinesin walks hand-over-hand},
  journal= {Science},
  volume = {303},
  pages  = {676--678},
  year   = {2004}
}

@book{Howard2001,
  author    = {Howard, J.},
  title     = {Mechanics of Motor Proteins and the Cytoskeleton},
  publisher = {Sinauer Associates},
  year      = {2001}
}

@book{Bray2001,
  author    = {Bray, D.},
  title     = {Cell Movements: From Molecules to Motility},
  edition   = {2},
  publisher = {Garland Science},
  year      = {2001}
}

@book{AlbertsMBoC7,
  author    = {Alberts, B. and Hopkin, K. and Johnson, A. D. and Morgan, D. and Roberts, K. and Walter, P.},
  title     = {Molecular Biology of the Cell},
  edition   = {7},
  publisher = {Garland Science},
  year      = {2022}
}

@book{LodishMCB9,
  author    = {Lodish, H. and Berk, A. and Kaiser, C. A. and Krieger, M. and Bretscher, A. and Ploegh, H. and Amon, A. and Scott, M. P.},
  title     = {Molecular Cell Biology},
  edition   = {9},
  publisher = {W. H. Freeman},
  year      = {2023}
}