18.04.02 · organismal-bio / musculoskeletal

Muscle contraction — the actin-myosin cycle

draft3 tiersLean: nonepending prereqs

Anchor (Master): Boron-Boulpaep advanced sections (Ch. 9 §III–V); Hill *Animal Physiology* 4th ed. (muscle and locomotion); Squire *Molecular Mechanism of Muscular Contraction* (Springer 1981) and Squire *The Structural Basis of Muscular Contraction* (Plenum 1981); primary literature — Huxley & Hanson 1954, Hill 1938, Lymn & Taylor 1971, Huxley & Simmons 1971, Rayment et al. 1993, Finer-Simmons-Spudich 1994

Intuition [Beginner]

A muscle is the body's mechanical engine. When you decide to lift a cup, the brain sends an electrical signal down a motor nerve, the signal arrives at a muscle, and the muscle shortens — pulling on the bone it is attached to, which moves the cup. Inside the muscle, this shortening is produced by millions of tiny protein machines working in parallel, each one taking a sliding step. The whole muscle shortens because the steps add up.

The two proteins doing the work are called actin and myosin. Actin assembles into long thin filaments; myosin assembles into thicker filaments with little projecting heads, like a bottle brush with the bristles pointing at the actin. The myosin heads are the moving parts. Each head can grab an actin filament, pull on it, let go, and grab again — over and over. Each pull is a small motion, on the order of a few nanometres, but billions of heads pulling at once on filaments arranged in parallel inside a muscle fibre produce a force you can feel.

Crucially, neither filament shortens. The two filaments overlap, and when the muscle contracts the myosin heads drag the actin filaments deeper into the overlap, like two interlocking combs sliding closer together. This is the sliding-filament model, the central organising idea of muscle contraction, proposed simultaneously in 1954 by Huxley and Hanson and by A. F. Huxley and Niedergerke. From electron-microscope pictures of muscle, taken at the time, you can see distinct bands — light and dark stripes — which become longer or shorter as the muscle stretches or contracts. The dark bands (where myosin is) stay the same width. The light bands (actin alone) shrink. Only the overlap zone changes. The filaments slide; they do not shrink.

Three more facts complete the picture.

First, calcium is the trigger. At rest, the myosin heads cannot bind to actin because a regulatory protein on the actin filament (tropomyosin) is in the way. When the nerve signal arrives, calcium ions are released from a special store inside the muscle cell, the calcium binds to another regulatory protein (troponin), troponin pulls tropomyosin off the binding sites, and the myosin heads can attach. When the signal stops, calcium is pumped back into the store, tropomyosin slides back over the binding sites, and the muscle relaxes.

Second, ATP is the fuel. Each pull cycle of a myosin head requires the hydrolysis of one molecule of ATP — the body's energy currency — into ADP and inorganic phosphate. The chemical energy released is converted into the mechanical work of the pull. This is why dead muscle is stiff (no fresh ATP to release the heads from actin: rigor mortis), and why a muscle that has run out of ATP cannot contract further or relax.

Third, one cycle has roughly five steps, repeated billions of times across the muscle. You can think of the cycle as a hand-over-hand walk along a rope: the myosin head binds the rope, pulls itself along, releases, re-cocks itself, binds again further down. The five steps in order: a myosin head holding an ATP is detached and floating; the ATP gets split (hydrolyzed) and the head re-cocks into an armed position; the armed head binds weakly to actin; phosphate is released and the binding strengthens, the head swings (this is the power stroke that produces force); ADP is released and the head is now tightly bound. A new ATP arrives, the head releases actin, and the cycle repeats.

The macroscopic muscle inherits these features. There are three distinct kinds of muscle: skeletal muscle (the voluntary kind that moves bones — striped under the microscope because the filaments are organised into regular bands), cardiac muscle (the involuntary, striped muscle of the heart, with extra coupling between cells), and smooth muscle (the involuntary, unstriped muscle of blood vessels, gut, airways — different regulation of calcium and a different molecular layout, but the same actin-myosin core). All three exploit the same actin-myosin cycle; they differ in how the cycle is regulated, how the filaments are arranged, and how the cells are wired together.

The headline of the unit: contraction is not a mystery substance that flows through a cable. It is a parallel array of small chemical machines, each of which converts the energy of an ATP molecule into a small mechanical pull, summed over billions of machines to produce force at the scale of a muscle.

Visual [Beginner]

Picture a muscle cell — a long thin fibre — under an electron microscope. The interior is divided into many parallel cylindrical strands called myofibrils, each one organised into a repeating pattern that gives the cell its striped (striated) appearance. The repeating unit is the sarcomere, roughly $2 μ m$ long at rest, the fundamental contractile unit of muscle.

Inside a single sarcomere, two sets of filaments overlap. The thick filaments (myosin) sit in the middle of the sarcomere. The thin filaments (actin) anchor at both ends of the sarcomere and reach inward, partially overlapping the thick filaments. Picture two combs facing each other, teeth meeting between them, the combs anchored at the outer ends and meeting in the middle. The teeth — the myosin heads — reach across and grab the actin filaments.

When the muscle contracts, the myosin heads drag the actin filaments inward, so the two anchor points (the "ends" of the comb) move closer together. The combs themselves do not get shorter; the distance between the anchor points shrinks. Bands you would see in the electron microscope: a wide dark band in the middle (where the thick filaments are — the A-band), a lighter band on either side (where only actin reaches — the I-band), and a thin dark line at each end where the actin filaments anchor (the Z-disc). On contraction the I-band narrows; the A-band stays the same; the Z-discs move closer together.

Now zoom into a single myosin head, attached to an actin filament. Picture a hand grasping a rope. The hand has a wrist (a flexible neck region attached to the rest of the myosin molecule) and an angle: cocked back like a flexed wrist before the power stroke, swung forward like an extended wrist after the power stroke. Each cycle of the hand is a small flick of the wrist — a few nanometres of motion at the contact point. Billions of hands flicking at slightly different times, on filaments arranged in parallel, sum to a smooth, continuous contraction at the macroscopic scale.

Worked example [Beginner]

A skeletal muscle fibre is at rest, with sarcomeres at length $2.0 μ m$ . A motor neuron fires, triggering contraction. We walk through the events in time.

Step 1. The action potential from the motor neuron arrives at the neuromuscular junction. Acetylcholine is released into the synaptic gap, opens ion channels on the muscle membrane, and a new action potential propagates along the muscle fibre and into the T-tubules (membrane invaginations that carry the signal deep into the cell).

Step 2. The action potential triggers the release of calcium from the sarcoplasmic reticulum (the internal calcium store wrapped around the myofibrils). Within milliseconds, intracellular calcium concentration rises from about $0.1 μ M$ at rest to about $10 μ M$ , a hundred-fold increase.

Step 3. Calcium binds to troponin on the actin filaments. The troponin-tropomyosin complex shifts, exposing the myosin binding sites on actin. The muscle is now "armed".

Step 4. The myosin heads (already in their cocked, ATP-hydrolyzed state from the previous resting cycle) bind weakly to actin. Phosphate dissociates, the binding strengthens, the heads swing through their power stroke. Each swing moves an actin filament by about $5 nm$ relative to the myosin filament — pulling the actin filament inward by this distance.

Step 5. ADP dissociates from each head; new ATP binds; the heads detach from actin. The fresh ATP is hydrolyzed (still on the head, before re-binding), and the heads re-cock into the armed position, ready for the next attachment. While calcium remains high, this cycle repeats — each head attaching, swinging, detaching, re-cocking — perhaps $5 - 20$ times per second per head, with different heads in different phases at any moment so the macroscopic force is smooth.

Step 6. After several cycles, the sarcomere has shortened from $2.0 μ m$ to (say) $1.6 μ m$ — a $20%$ contraction. Across the whole muscle fibre (with maybe $10, 000$ sarcomeres in series), this scales to a $20%$ shortening of the fibre as a whole. A $10 cm$ muscle shortens by $2 cm$ — enough to flex a joint.

Step 7. The motor neuron stops firing. Calcium is pumped back into the sarcoplasmic reticulum by a calcium-ATPase (an ATP-consuming pump). Intracellular calcium drops; troponin releases calcium; tropomyosin slides back over the myosin binding sites; new myosin attachments cannot form; the existing attached heads complete their current cycle and detach. The muscle relaxes.

What this tells us: the contraction is built from a single repeating cycle (steps 4 and 5) running in parallel on billions of heads, gated by the calcium signal that opens and closes the actin binding sites. Each cycle consumes one ATP; ATP consumption is the price of the mechanical work. A muscle that has run out of ATP can neither contract further (no cycling) nor relax (heads cannot detach without fresh ATP — this is the molecular origin of rigor mortis).

Check your understanding [Beginner]

Exercise (easy, multiple choice).

During muscle contraction, which of the following actually shortens?

A. The thick (myosin) filaments B. The thin (actin) filaments C. The distance between the Z-discs of each sarcomere D. The myosin heads

Hint

The sliding-filament model says the filaments slide past each other; they do not change length.

Answer

The distance between the Z-discs of each sarcomere — option C.

The thick and thin filaments slide past each other; neither one shortens. The Z-discs (the anchor points of the thin filaments at the ends of each sarcomere) move closer together because the myosin heads drag the actin filaments inward. The H-zone (the central region where only myosin sits) and the I-band (where only actin sits) both shrink during contraction; the A-band (where the thick filaments are) stays the same width.

Formal definition [Intermediate+]

Sarcomere architecture

A sarcomere is the structural and functional unit of striated muscle, the region between two adjacent Z-discs (also called Z-lines) along a myofibril. At resting length in mammalian skeletal muscle, a sarcomere is about $2.0 - 2.5 μ m$ long. Standard nomenclature, observed by electron microscopy and X-ray diffraction:

Z-disc — the lateral boundary of the sarcomere; an electron-dense plate of $α$ -actinin and associated proteins where the thin (actin) filaments anchor with their barbed ( $+$ ) ends pointing into the sarcomere.
I-band — the region adjacent to a Z-disc containing only thin filaments. Its width changes with sarcomere length.
A-band — the central region of the sarcomere, the full length of the thick (myosin) filaments. Its width is fixed at the thick-filament length, approximately $1.6 μ m$ .
H-zone — the central portion of the A-band where only thick filaments sit (no actin overlap). Its width changes with sarcomere length.
M-line — a thin dense region at the centre of the H-zone, the protein scaffold (myomesin and others) that cross-links the thick filaments laterally.

Length changes during contraction: the A-band is invariant; the H-zone and I-band both shrink; the distance between Z-discs (sarcomere length) shrinks. The thick and thin filaments themselves are length-invariant in this picture.

The thin filament is a double helix of polymerised G-actin monomers (resulting in F-actin), decorated with the regulatory complex tropomyosin (a coiled-coil $α$ -helical dimer covering the actin grooves) and troponin (a heterotrimer Tn-C, Tn-I, Tn-T positioned every seven actin monomers).

The thick filament is an assembly of $\sim 300$ myosin II molecules, arranged with their tail domains forming the central bare-zone region of the filament (a bipolar structure with opposite polarity on either side of the M-line) and their head domains projecting outward in a regular helical array. Each myosin II molecule is a hexamer: two heavy chains (each containing the motor head domain, the lever-arm neck, and the coiled-coil tail) plus four light chains (two essential, two regulatory). The motor head plus neck is the proteolytic fragment called subfragment-1 (S1); head plus neck plus part of the coiled-coil tail is S2.

The cross-bridge cycle (five-state)

A single myosin head undergoes a cyclic sequence of chemical states correlated with mechanical states. The standard five-state nomenclature, derived from the Lymn-Taylor 1971 kinetic scheme ^{[Lymn-Taylor 1971]} extended with the rapid hydrolysis step:

$M \cdot A T P$ (detached, post-rigor): a free (actin-detached) myosin head with bound ATP. Low affinity for actin.
$M \cdot A D P \cdot P_{i}$ (detached, cocked): ATP has been hydrolyzed to ADP + $P_{i}$ , both still bound to the head. The head has rotated into the "cocked" or "pre-power-stroke" position. The hydrolysis step is fast and freely reversible; the products remain in the active site.
$A M \cdot A D P \cdot P_{i}$ (weakly attached): the cocked head has bound actin (denoted $A$ ) but in a weak, pre-power-stroke conformation. This step is rate-limited by the search for an available binding site on actin (calcium-gated).
$A M \cdot A D P$ (strongly attached, post-power-stroke): phosphate has been released. Phosphate release is tightly coupled to the power stroke — a rotation of the lever arm relative to the actin-binding domain, displacing the actin filament by about $5 - 10 nm$ relative to the thick filament.
$A M$ (rigor): ADP has dissociated. The head is tightly bound to actin with no nucleotide.

The cycle closes via:

$A M + A T P \to M \cdot A T P$ : a new ATP binds the rigor complex, rapidly weakening the actin affinity and releasing the head (this is the step that fails in rigor mortis: no ATP, no release).
Return to step 2 by re-hydrolyzing the new ATP.

The cycle is driven thermodynamically by ATP hydrolysis: the chemical free-energy difference $Δ μ_{ATP} = Δ G^{0} - R T ln ([ATP] / ([ADP] [P_{i}])) \approx - 50 kJ mo l^{- 1}$ at physiological concentrations. Each completed cycle consumes one ATP. The cycle is directional because the strongly-bound states (4, 5) preferentially form after the weak states (3) — chemistry of phosphate release is mechanically coupled to the lever-arm rotation in one direction.

Excitation-contraction coupling

Skeletal muscle is calcium-regulated at the thin filament. At rest, intracellular $[Ca^{2 +}]_{i} \approx 0.1 μ M$ and tropomyosin sterically blocks the myosin binding sites on actin (the "blocked" or "off" state of the thin filament). Activation propagates as follows:

An action potential in the sarcolemma propagates into the T-tubules.
T-tubular depolarisation activates dihydropyridine receptors (Ca $_{v}$ 1.1 in skeletal muscle), which mechanically gate the type-1 ryanodine receptor (RyR1) on the sarcoplasmic reticulum (SR) membrane.
RyR1 opens; calcium released from the SR raises $[Ca^{2 +}]_{i}$ to $\sim 10 μ M$ within $\sim 5 ms$ .
$Ca^{2 +}$ binds troponin C; conformational change propagates through troponin I and troponin T; tropomyosin moves azimuthally on the actin filament by $\sim 2 5^{\circ}$ , uncovering the myosin binding sites.
The cross-bridge cycle proceeds as above.
Calcium is removed from the cytoplasm by SERCA (the sarcoplasmic-reticulum calcium-ATPase) — itself an ATP-consuming pump — back into the SR. As $[Ca^{2 +}]_{i}$ falls, calcium dissociates from troponin C, tropomyosin returns to the blocking position, and the muscle relaxes.

Cardiac muscle uses the same RyR/SR mechanism but with calcium-induced calcium release: a small trans-sarcolemmal $Ca^{2 +}$ influx through L-type channels triggers RyR2 opening (chemical rather than mechanical coupling). Smooth muscle uses an alternative regulatory pathway with myosin-light-chain phosphorylation (calmodulin + myosin light-chain kinase) — regulation is on the thick filament rather than the thin filament, and there is no organised sarcomere.

The Hill 1938 force-velocity relation

For a tetanically stimulated muscle shortening against an isotonic load $F$ , the steady-state shortening velocity $v$ obeys the Hill equation ^{[Hill 1938]}

(F + a) (v + b) = (F_{0} + a) b,

where $F_{0}$ is the isometric tension (the maximum tension developed when $v = 0$ ), and $a, b$ are empirical constants with units of force and velocity respectively. The ratio $a / F_{0}$ is dimensionless and typically takes values $0.15 - 0.30$ for vertebrate skeletal muscle; $b / v_{m a x}$ has the same value when $v_{m a x}$ is the unloaded shortening velocity, $v_{m a x} = b F_{0} / a$ .

Equivalent rearranged forms:

v (F) = \frac{b ( F _{0} - F )}{F + a}, F (v) = \frac{F _{0} b - a v}{v + b} .

The curve is hyperbolic in the $(F, v)$ plane, asymptotic to $F = - a$ at large $v$ and to $v = - b$ at large $F$ (the asymptotes lie outside the physiological quadrant). It interpolates between two limits: maximum force at zero shortening velocity (isometric, $F = F_{0}$ at $v = 0$ ) and maximum shortening velocity under no load (unloaded, $F = 0$ at $v = v_{m a x} = b F_{0} / a$ ).

The length-tension relation

The active tension developed by a tetanized fibre at fixed length depends on the geometry of thick-thin filament overlap. The maximum active tension occurs at the optimal sarcomere length $L_{0} \approx 2.0 - 2.2 μ m$ in mammalian skeletal muscle, where every myosin head along the thick filament has an actin filament available to bind. Both shorter and longer sarcomere lengths reduce tension — at longer lengths because fewer heads can reach actin (overlap shrinks); at shorter lengths because thin filaments from opposing sides interfere and because the thick filament collides with the Z-disc. The qualitative curve has a plateau between $\sim 2.0$ and $\sim 2.2 μ m$ (where the bare zone is uncovered by actin), descending arms on either side, and predicted slopes that match the measured filament dimensions.

This length-tension curve was the original quantitative evidence for the sliding-filament model: it correlates active tension with the geometric overlap that the sliding-filament picture predicts, and rules out competing models (e.g., contractile-protein-shortening or solation-gelation models) that have no such geometric prediction.

Counterexamples to common slips

The filaments themselves do not shorten. A common Beginner-level error is to picture actin or myosin contracting. Both filaments are length-invariant in the sliding-filament picture; only their overlap zone changes, and both myosin heads and thin-filament length are invariants of the cycle.
The myosin power stroke is not "pull-then-release". It is "weakly-bind, then phosphate-release-coupled lever-arm rotation, then ADP-release, then ATP-rebinding releases the head." The lever-arm rotation occurs while attached; the head releases actin only after a new ATP arrives.
The "myosin step size" is not the same as the "lever-arm rotation". The lever arm rotates by a large angle ( $\sim 60 - 7 0^{\circ}$ between pre- and post-stroke states from X-ray structures), but the contact point on actin moves by only $5 - 10 nm$ — the geometry of the lever arm converts a large angular rotation into a small linear displacement at the actin filament.
Rigor is not the same as the "strongly bound" state of the cycle. Rigor is the nucleotide-free state $A M$ ; the strongly-bound post-power-stroke state $A M \cdot A D P$ is one step before rigor. Both are tightly actin-bound; only rigor lacks any nucleotide.
Hill's $a$ is not a friction coefficient. Despite the linear-friction-like appearance of $F + a$ as a denominator, $a$ has units of force and arises in Hill's derivation as a heat-of-shortening constant (the rate of additional heat liberated per unit shortening). Modern microscopic derivations (Huxley 1957 rate-distortion model) identify $a$ with a cross-bridge-cycling parameter.
Cardiac muscle is not "skeletal muscle that contracts continuously". Cardiac muscle has distinct gene expression (different myosin heavy-chain isoforms, $β$ -MHC dominant in human ventricle), calcium-induced-calcium-release (chemical rather than mechanical coupling), gap junctions for intercellular electrical coupling at intercalated discs, and length-dependent activation (sarcomere length modulates myofilament calcium sensitivity — the cellular basis of the Frank-Starling mechanism).

Key theorem with proof [Intermediate+]

We derive the Hill 1938 force-velocity relation from the Huxley 1957 cross-bridge model. This is the canonical derivation linking molecular kinetics to macroscopic muscle mechanics — and it is the foundational result of muscle biophysics.

Theorem (Hill force-velocity from Huxley cross-bridge kinetics). Consider a population of cross-bridges, each described by its position $x$ along the actin filament relative to its equilibrium attachment point on the thick filament. Let $n (x, t)$ denote the fraction of cross-bridges attached at position $x$ at time $t$ , and assume:

(i) The thin filament slides relative to the thick filament with velocity $v$ (positive $v$ = shortening), so each attached cross-bridge's position evolves by $d x / d t = - v$ .

(ii) Cross-bridges attach at rate $f (x)$ (only when unattached) and detach at rate $g (x)$ , with $f$ supported on $x > 0$ and $g$ supported on the whole line.

(iii) Each attached cross-bridge contributes a linear-spring force $k x$ along the filament direction, with $k$ the cross-bridge stiffness.

(iv) The system is in steady state at fixed $v$ ( $\partial n / \partial t = 0$ ).

Then the steady-state population $n (x; v)$ satisfies

- v \frac{\partial n}{\partial x} = f (x) [1 - n (x)] - g (x) n (x),

and for the Huxley 1957 piecewise-linear choice $f (x) = f_{1} \cdot min (x / h, 1) 1_{x > 0}$ and $g (x) = g_{2} \cdot 1_{x < 0} + g_{1} \cdot (x / h) 1_{0 < x < h}$ , the total force $F (v) = N \int k x n (x; v) d x$ is a hyperbolic function of $v$ of the Hill form

(F + a) (v + b) = (F_{0} + a) b

with $F_{0} = N k h^{2} f_{1} / [2 (f_{1} + g_{1})]$ , $b = h (f_{1} + g_{1}) / g_{2}$ (in appropriate normalisation of the units), and $a / F_{0}$ a function of the rate ratios $g_{2} / (f_{1} + g_{1})$ .

Proof. (Steady-state equation.) The cross-bridge population is governed by the PDE

\frac{\partial n}{\partial t} + \frac{d x}{d t} \frac{\partial n}{\partial x} = f (x) [1 - n] - g (x) n .

Substituting $d x / d t = - v$ and the steady-state assumption $\partial n / \partial t = 0$ :

- v \frac{\partial n}{\partial x} = f (x) [1 - n] - g (x) n . (*)

For the Huxley-1957 piecewise-linear rates, equation ( $*$ ) is a first-order linear ODE in $x$ at each fixed $v$ . On the active region $0 < x < h$ , $f (x) = f_{1} x / h$ and $g (x) = g_{1} x / h$ , so

v \frac{d n}{d x} = (f_{1} + g_{1}) (x / h) n - f_{1} (x / h),

with steady-state solution (after imposing the boundary condition $n (h) =$ value from the matched region $x > h$ , where $f = 0$ , and assuming $n \to 0$ as $x \to \infty$ for shortening $v > 0$ ):

n (x; v) = \frac{f _{1}}{f _{1} + g _{1}} [1 - exp (- (f_{1} + g_{1}) (h^{2} - x^{2}) / (2 h v))],

valid on $0 < x < h$ . On $x < 0$ , $f = 0$ and $g = g_{2}$ , giving $n (x; v) = n (0; v) exp (g_{2} x / v)$ (an exponentially decaying tail). On $x > h$ , no attachment ( $f = 0$ ) and no detachment in the simplest model ( $g = 0$ on $x > h$ , or a small constant), giving the boundary-matching constraint.

(Force and the Hill form.) The total cross-bridge force per half-sarcomere is

F (v) = N k \int_{- \infty}^{\infty} x n (x; v) d x,

with $N$ the cross-bridge number density. Substituting the steady-state solution above and integrating, the dominant contribution from the active region yields

F (v) = \frac{N k h ^{2} f _{1}}{2 ( f _{1} + g _{1} )} Φ (\frac{( f _{1} + g _{1} ) h}{v}),

where $Φ$ is a smooth monotone function. At $v = 0$ , $Φ (\infty) = 1$ and $F (0) = N k h^{2} f_{1} / [2 (f_{1} + g_{1})] := F_{0}$ . At large $v$ , $Φ$ vanishes; the unloaded shortening velocity is $v_{m a x}$ at which the negatively-contributing tail $x < 0$ exactly cancels the positive contribution. The hyperbolic Hill form $(F + a) (v + b) = (F_{0} + a) b$ is recovered as a first-order rational approximation of $Φ$ — the precise identification of $a$ and $b$ with combinations of $f_{1}, g_{1}, g_{2}$ depends on which moment of the curve is matched.

Huxley's 1957 numerical fit to frog sartorius data using this model recovered force-velocity, length-tension, and heat-of-shortening curves with no free parameters beyond $f_{1}, g_{1}, g_{2}, h$ ; the fit was the original quantitative success of the cross-bridge picture, predating any direct structural evidence for the cross-bridge cycle.

Remark. The proof reduces a macroscopic mechanical law (Hill force-velocity) to a molecular-kinetic statement (steady-state of a reaction-diffusion equation on cross-bridge position) plus one geometric input (linear-spring cross-bridges with characteristic length $h$ ). This reduction is the foundational claim of cross-bridge theory: the constitutive law of muscle as a continuum is dictated by the cycle kinetics of its molecular machinery.

The Huxley-Simmons 1971 extension ^{[Huxley-Simmons 1971]} refines the model by replacing the single attached state with two discrete attached states (pre- and post-stroke) connected by a fast transition; this two-state attached version is what fits the rapid T1-T2 length-step transients observed in single-fibre experiments. ∎

Exercises [Intermediate+]

Exercise 1 (easy, short-answer).

List the five states of the Lymn-Taylor cross-bridge cycle in order, naming the chemical species bound at the myosin head (ATP, ADP, ADP·Pi, or none) and whether the head is attached to actin.

Hint

Start from the detached, post-hydrolysis state with the head in the cocked position.

Answer

$M \cdot A T P$ : detached, ATP bound, low actin affinity (post-rigor, pre-hydrolysis).
$M \cdot A D P \cdot P_{i}$ : detached, ADP + $P_{i}$ bound, head cocked into pre-power-stroke conformation. ATP hydrolysis has occurred; products remain in the active site.
$A M \cdot A D P \cdot P_{i}$ : weakly attached to actin, ADP + $P_{i}$ still bound. Calcium-gated step — only proceeds when tropomyosin has uncovered the actin binding site.
$A M \cdot A D P$ : strongly attached, post-power-stroke. Phosphate has been released; the lever arm has rotated.
$A M$ : rigor state, no nucleotide, tightly bound. Persists indefinitely without ATP.

The cycle closes by $A M + A T P \to M \cdot A T P$ (new ATP binds the rigor complex and releases the head), returning to state 1.

Exercise 2 (easy, short-answer).

Predict the qualitative effect of (a) removing all calcium from the extracellular and intracellular environments and (b) running the fibre out of ATP, on a previously contracting skeletal muscle fibre.

Hint

Calcium gates the attachment step (state 2 → state 3). ATP gates the detachment step (state 5 → state 1).

Answer

(a) Removing calcium: tropomyosin returns to the blocking position; no new myosin attachments form. Existing attached cross-bridges complete their current cycle and detach upon ATP rebinding (since ATP is still available); the fibre relaxes within milliseconds. This is the basis of normal relaxation in life.

(b) Removing ATP: existing cross-bridges complete their power stroke (state 4) and then release phosphate and ADP (state 5, rigor) but cannot detach because no fresh ATP arrives. New cycles cannot start. The fibre becomes stiff and immobile, locked in the rigor configuration with all heads tightly bound. This is rigor mortis at the molecular level.

The two manipulations are dual: calcium controls the "attach" gate (state 2 → 3), ATP controls the "release" gate (state 5 → 1). Loss of the attach gate gives relaxation; loss of the release gate gives rigor.

Exercise 3 (medium, numeric).

A skeletal muscle fibre has cross-sectional area $A = 1 0^{- 6} m^{2}$ ( $1 m m^{2}$ ) and generates a tetanic isometric tension of $F_{0} = 0.3 N$ . Estimating the cross-bridge density as $ρ = 2 \times 1 0^{16} m^{- 2}$ active heads per unit cross-sectional area, with a fraction $ϕ = 0.3$ of heads attached at any instant during isometric activation, compute the average force per attached cross-bridge in piconewtons.

Hint

Total force = (attached heads per area) × (force per head) × (area). Solve for force per head.

Answer

The number of attached heads in the cross-section is $N_{att} = ϕρ A = 0.3 \times 2 \times 1 0^{16} m^{- 2} \times 1 0^{- 6} m^{2} = 6 \times 1 0^{9}$ heads.

Force per head: $F_{head} = F_{0} / N_{att} = 0.3 N / (6 \times 1 0^{9}) = 5 \times 1 0^{- 11} N = 50 pN$ .

The order-of-magnitude answer is in the few-pN to tens-of-pN range, consistent with single-molecule laser-trap measurements that have given $3 - 10 pN$ per cross-bridge depending on the experimental conditions and how "force per cross-bridge" is normalised against series compliance. The detailed reconciliation between bulk-muscle and single-molecule force measurements was a major effort of the 1990s.

Exercise 5 (medium, symbolic).

The length-tension curve of skeletal muscle (Gordon-Huxley-Julian 1966) has a plateau from sarcomere length $L_{1} \approx 2.0 μ m$ to $L_{2} \approx 2.2 μ m$ at peak tension $F_{0}$ , a descending arm from $L_{2}$ to about $3.65 μ m$ where active tension reaches zero. Given that the thick filament is $1.6 μ m$ long and each half of a thin filament is approximately $1.0 μ m$ long, and that there is a $0.2 μ m$ central bare zone on the thick filament, predict $L_{1}, L_{2}$ , and the zero-tension length on the descending arm.

Hint

The plateau corresponds to lengths where the bare zone is fully covered by neither thin filament. The zero-tension length is where the overlap between thick and thin filament is exactly zero.

Answer

Let the sarcomere length be $L$ , thick filament length be $T = 1.6 μ m$ , half-thin-filament length be $A_{h} = 1.0 μ m$ , bare zone length be $B = 0.2 μ m$ .

The overlap between the thick filament and one thin filament is $max (0, min (T /2, A_{h}, A_{h} - (L /2 - T /2)))$ on each side of the M-line. The cross-bridge count engaged scales with the overlap minus the part of the thick filament that falls inside the bare zone (where there are no heads).

Plateau condition: every head along the thick-filament-minus-bare-zone region has actin available, i.e., $(L - T) /2 \leq A_{h} - B /2$ , which gives $L \leq T + 2 (A_{h} - B /2) = 1.6 + 2 (1.0 - 0.1) = 3.4 μ m$ as the upper edge of the plateau ... wait, this needs care. Let me redo more carefully: the plateau is the range where (i) no overlap onto the bare zone reduces head-recruitment, (ii) no thin-filament collision from the opposite half-sarcomere occurs.

The plateau upper edge $L_{2}$ is the length at which the actin tips just reach the edge of the bare zone, giving $L_{2} = 2 A_{h} + B = 2.0 + 0.2 = 2.2 μ m$ . The plateau lower edge $L_{1}$ is the length at which actin from one half-sarcomere just meets actin from the other half: $L_{1} = 2 A_{h} = 2.0 μ m$ . The zero-tension descending-arm length is $L_{0} = T + 2 A_{h} = 1.6 + 2.0 = 3.6 μ m$ , matching the measured value $\sim 3.65 μ m$ to within the precision of filament-length measurement.

The match between predicted and measured length-tension is the original quantitative evidence for the sliding-filament hypothesis and rules out competing models (e.g., contractile-protein shortening) that have no geometric prediction for the descending arm.

Exercise 6 (medium, short-answer).

Compare excitation-contraction coupling in skeletal vs cardiac vs smooth muscle along three axes: (a) signal that initiates calcium release from internal stores; (b) regulatory protein on which calcium acts; (c) whether the calcium-binding step is on the thin or thick filament.

Hint

Skeletal: mechanical coupling. Cardiac: chemical coupling. Smooth: phosphorylation.

Answer

	Skeletal	Cardiac	Smooth
(a) Signal for SR release	Mechanical coupling: DHPR (Ca $_{v}$ 1.1) on T-tubule directly gates RyR1 on SR. No extracellular Ca needed for the gating itself.	Chemical: extracellular Ca $^{2 +}$ enters via L-type Ca $^{2 +}$ channels (DHPR Ca $_{v}$ 1.2) and triggers RyR2 opening by calcium-induced calcium release.	Less SR; some intracellular release via IP $_{3}$ receptors and RyR3; major component is extracellular Ca influx via L-type channels and store-operated channels.
(b) Ca-binding regulatory protein	Troponin C on the thin filament.	Cardiac troponin C (similar to skeletal but with isoform-specific Ca-binding affinities).	Calmodulin in the cytoplasm — not on a filament.
(c) Thin or thick filament regulation	Thin filament — Ca shifts tropomyosin off myosin binding sites.	Thin filament — same mechanism as skeletal.	Thick filament — Ca-calmodulin activates myosin light-chain kinase (MLCK) which phosphorylates the regulatory myosin light chain, activating myosin ATPase.

Smooth-muscle regulation is fundamentally different — the regulatory step is a covalent modification (phosphorylation) of the thick filament rather than a steric un-blocking of the thin filament. Smooth muscle also lacks organised sarcomeres; its actin and myosin filaments form a less regular contractile network anchored to dense bodies in the cytoplasm and to membrane plaques on the sarcolemma.

Exercise 7 (hard, symbolic).

For a population of $N$ cross-bridges, each binding actin with rate $f$ , detaching with rate $g$ , and stochastically contributing a force $F_{1}$ per attached bridge, derive the mean and variance of the total force in steady state. Identify the regime in which the relative force fluctuation $σ_{F} / ⟨ F ⟩$ becomes small enough that a "macroscopic" muscle behaves as a smooth, deterministic system rather than a fluctuating molecular ensemble.

Hint

In steady state, the number of attached cross-bridges follows a binomial distribution with attachment probability $p = f / (f + g)$ and $N$ trials. Use the binomial mean and variance.

Answer

In steady state, each cross-bridge is independently attached with probability $p = f / (f + g)$ and detached with probability $1 - p = g / (f + g)$ . The number $n$ of attached bridges is binomial with parameters $N$ and $p$ , so

⟨ n ⟩ = N p, Var (n) = N p (1 - p) .

Total force $F = F_{1} n$ , so

⟨ F ⟩ = F_{1} N p, Var (F) = F_{1}^{2} N p (1 - p), σ_{F} = F_{1} N p (1 - p) .

Relative fluctuation:

\frac{σ _{F}}{⟨ F ⟩} = \frac{1}{N} \frac{1 - p}{p} .

This scales as $1/ N$ . For a single half-sarcomere with $N \sim 1 0^{2}$ active cross-bridges, fluctuations are $\sim 10%$ — observable in single-fibre experiments. For a whole muscle fibre with $N \sim 1 0^{10}$ cross-bridges acting in parallel, fluctuations are $\sim 1 0^{- 5}$ — utterly invisible at the macroscopic scale. This is the law-of-large-numbers reason a macroscopic muscle behaves as a smooth deterministic continuum despite being built from stochastic molecular machines.

The numerical observability of cross-bridge fluctuations at the single-fibre level is the historical reason single-molecule and single-fibre experiments (Finer-Simmons-Spudich 1994) were able to resolve the discrete cycle structure; bulk-muscle force-velocity data alone are smoothed by $N$ -averaging and cannot resolve individual events.

Exercise 8 (hard, numeric).

A muscle does $W = 1 J$ of mechanical work during a sustained contraction. The free energy of ATP hydrolysis is $Δ μ_{ATP} \approx - 50 kJ mo l^{- 1}$ under physiological conditions. The muscle consumes $5 \times 1 0^{19}$ molecules of ATP during this contraction. Compute the thermodynamic efficiency $η = W / (- Δ G_{ATP consumed})$ .

Hint

Convert ATP molecules to moles via Avogadro's number, then to free energy in joules.

Answer

Moles of ATP consumed: $n_{ATP} = 5 \times 1 0^{19} / (6.022 \times 1 0^{23}) \approx 8.3 \times 1 0^{- 5} mol$ .

Free energy released: $∣Δ G ∣ = n_{ATP} \times 50, 000 J mo l^{- 1} \approx 4.15 J$ .

Efficiency: $η = W /∣Δ G ∣ = 1 J /4.15 J \approx 0.24$ , i.e., about $24%$ .

This is in the right range — measured efficiencies of vertebrate skeletal muscle running near optimal load are $20 - 40%$ depending on conditions, with cardiac muscle in the upper part of that range under optimal load. The remaining $\sim 75%$ is dissipated as heat (the rest of the "heat-of-shortening" that Hill measured calorimetrically in 1938). The thermodynamic upper bound for a chemical motor working between $ATP$ and $ADP + P_{i}$ chemical potentials is essentially $100%$ in principle (the cycle is not heat-bath-limited like a Carnot engine); the measured $20 - 40%$ reflects cycle-internal dissipation (each step is finite-rate, not quasi-static) and the cost of running the cycle against the resisting load.

Exercise 9 (hard, short-answer).

A single-molecule laser-trap experiment (Finer-Simmons-Spudich 1994 style) records the displacement of an actin filament held between two beads, with a single myosin head tethered to a coverslip in the gap. The displacement trace shows discrete steps of $\sim 11 nm$ at random intervals, with each step lasting $\sim 50 ms$ before the system returns to baseline. Interpret the step size, the dwell time, and what each event represents in terms of the cross-bridge cycle. What would change if you decreased the ATP concentration tenfold?

Hint

The step is the power stroke. The dwell time at the displaced position is the attached lifetime. ATP concentration controls one specific transition in the cycle.

Answer

Each $11 nm$ step is the power stroke of a single myosin head: phosphate release coupled to lever-arm rotation displaces the actin filament by the working stroke length, with $5 - 15 nm$ being the typical range observed across experimental conditions and myosin isoforms. The dwell at the displaced position is the attached-state lifetime, i.e., the time the head spends in states $A M \cdot A D P$ and $A M$ before a new ATP binds and releases the head. Returning to baseline corresponds to head detachment (and the actin filament drifting back to its trap-centred position).

Decreasing $[ATP]$ tenfold prolongs the rigor-state dwell, because the rate of new ATP binding is first-order in $[ATP]$ (rate constant $k_{+ ATP} \times [ATP]$ ). Specifically, the detachment-step rate from state $A M$ scales linearly with $[ATP]$ , so the attached dwell time at high $[ATP]$ is limited by ADP release (slower step) and at low $[ATP]$ becomes limited by ATP binding. Quantitatively, at $1 mM$ ATP and typical rate constants $k_{+ ATP} \approx 1 0^{6} M^{- 1} s^{- 1}$ , the ATP-binding-limited dwell is $\sim 1 ms$ — negligible compared to a $50 ms$ ADP-release-limited dwell. At $0.1 mM$ ATP, ATP-binding-limited dwell becomes $\sim 10 ms$ , starting to be comparable. At sub- $μ M$ ATP, the dwell extends to seconds — approaching the rigor-state regime.

This kind of dwell-time analysis is the experimental backbone of modern cross-bridge kinetics: by tuning $[ATP]$ , $[ADP]$ , $[P_{i}]$ , and load, one can selectively prolong specific transitions in the cycle and assign rate constants to individual steps.

Lean formalization [Intermediate+]

Mathlib does not cover biology, and the relevant mathematical objects underlying muscle contraction — continuous-time Markov chains on a labelled state space with chemical-potential drive, the Huxley 1957 rate-distortion partial differential equation, the Hill 1938 force-velocity relation as a derived constitutive law — are not present in Mathlib in a form usable for a formal statement.

The pathway to formalisation is staged:

Continuous-time Markov chains as a Mathlib structure (closest current layer: Mathlib.Probability.ProbabilityMassFunction and discrete Markov chain material; nothing for the continuous-time generator on a labelled state space at the needed generality).
Driven non-equilibrium cycles (thermodynamic-cycle structure where one transition is biased by a chemical-potential difference; no current Mathlib analog).
The Huxley 1957 reaction-rate-PDE on cross-bridge density — a first-order linear PDE with position-dependent rates. Needs the basic theory of linear PDEs on $R$ and steady-state Green's function methods.
The Hill force-velocity relation as a corollary of the steady-state population integral.
The Huxley-Simmons 1971 two-state attached model and its T1-T2 transient response under length-step inputs.

lean_status: none reflects this absence. The unit is reviewer-attested. The aggregated lean_status: none units in §17–19 form a small portion of the upstream Mathlib biology-and-biophysics contribution roadmap; the load-bearing prerequisite is the continuous-time Markov chain / non-equilibrium driven cycle pair, which is foundational for many biology and chemistry units beyond muscle.

Single-molecule mechanics and the 1990s structural revolution [Master]

By the late 1980s the cross-bridge cycle was kinetically characterised (Lymn-Taylor 1971, Huxley-Simmons 1971) and the sliding-filament model was three decades old, but the atomic structures of the proteins and the single-molecule mechanics of one cross-bridge were both unresolved.

Two parallel developments closed this gap.

The Rayment 1993 structures. The X-ray crystal structure of the chicken skeletal myosin S1 fragment (subfragment-1: motor head plus lever-arm neck) ^{[Rayment 1993]} revealed the catalytic site at the nucleotide-binding cleft, the actin-binding interface on the opposite face of the head, and the converter-arm-lever architecture by which a small ( $\sim 1 nm$ ) reorganisation at the nucleotide-binding cleft is amplified into a large angular rotation ( $\sim 60 - 7 0^{\circ}$ ) of the lever arm. The companion paper on actin-myosin docking proposed the structural identification of the strong-binding interface. The lever-arm-amplification picture immediately rationalised the cycle-stepping data and made specific predictions: longer-lever-arm myosins (myosin VI with truncated lever, myosin V with elongated lever) should give different step sizes — predictions confirmed by single-molecule experiments through the 1990s and 2000s. The 1993 structure was a structural completion of the cycle whose kinetics had been worked out two decades earlier.

The Finer-Simmons-Spudich 1994 laser-trap. A single myosin head, tethered to a coverslip, attached to an actin filament suspended in a dual-laser trap. The trap held one bead at each end of the filament; head attachment displaced the filament against the trap stiffness; calibrated displacement and force readouts gave (i) the step size of a single power stroke (originally reported $\sim 11 nm$ for skeletal myosin II) and (ii) the force per attached cross-bridge ( $\sim 3 - 5 pN$ at low loads, with subsequent refinement up to $\sim 10 pN$ when series-compliance was accounted for). This was the first direct measurement at the single-cross-bridge level, decades after the cross-bridge cycle had been inferred indirectly from bulk-muscle data.

Subsequent single-molecule work (Yanagida, Block, Spudich, Goldman labs) refined the measurements and probed additional cycle features — the substep structure within a power stroke, the load-dependence of dwell times, the difference between processive (myosin V, kinesin) and non-processive (myosin II) motors. The latter distinction is structural: processive motors have two heads working in coordinated alternation so the molecule never fully releases its track, whereas non-processive motors release after every cycle and depend on parallel arrays of independent heads to sustain motion. The actin-myosin II system underlying muscle is non-processive by design — a muscle works because billions of heads cycle independently with overlapping duty cycles, each head making one stroke per cycle and detaching, with the population sustaining smooth force.

Thermodynamics of the cycle and stochastic mechanochemistry [Master]

The cross-bridge cycle is a driven non-equilibrium cycle: the chemical potential difference $Δ μ_{ATP} = μ_{ATP} - μ_{ADP} - μ_{P_{i}}$ is held fixed by metabolic machinery (creatine-phosphate buffering, oxidative phosphorylation, glycolysis) at a value $\approx - 50 kJ mo l^{- 1}$ at physiological substrate concentrations, making the cycle proceed unidirectionally despite each individual transition being microscopically reversible (microscopic-reversibility / detailed-balance is broken only at the cycle level by the chemical-potential drive).

Modeling the cycle as a continuous-time Markov chain on the state space ${1, 2, 3, 4, 5}$ (the five Lymn-Taylor states) with rate matrix $W_{ij}$ and chemical-potential-biased rates per the local-detailed-balance condition $W_{ij} / W_{j i} = exp (Δ s_{ij} / k_{B})$ (where $Δ s_{ij}$ is the entropy production of the $i \to j$ transition including ATP-cycle work), the steady-state distribution carries a net cycle flux. The flux is mechanically expressed as the cycle rate (one ATP per cycle, one step per cycle) and thermodynamically as the entropy production rate $\dot{Σ} = J_{cycle} \cdot Δ μ_{ATP} / T$ .

Efficiency, defined as $η = W_{out} /∣Δ G_{ATP consumed} ∣$ , has measured values $\sim 0.2 - 0.4$ in vertebrate skeletal muscle at near-optimal loads. Theoretical upper bounds: a strictly isothermal driven cycle has no Carnot-style temperature-based bound; the relevant bound is the Onsager limit at small affinities (linear-response regime) and the near-equilibrium limit $η \to 1$ for an infinitely slow cycle. Real muscle operates far from quasi-static; each transition runs at finite rate to deliver useful power at biological timescales, and the deviation from unit efficiency reflects this trade-off.

The modern formal apparatus is the stochastic thermodynamics of molecular motors: see Seifert (2012 review Rep. Prog. Phys. 75 126001), Astumian (1997 Science 276 917), and the textbook treatment in Phillips et al. Physical Biology of the Cell. The framework predicts, among other things, the efficiency-power trade-off (high efficiency requires near-quasi-static cycling, which has low power; high power requires near-stall conditions, which has low efficiency); the optimal operating point for a motor of given chemical drive; the noise spectrum and fluctuation-dissipation relations on the cycle flux.

The connection to the broader molecular-motor zoo — myosin (linear motor on actin), kinesin and dynein (linear motors on microtubules), $F_{1}$ ATP-synthase (rotary motor) — is that all share the driven-Markov-cycle framework with motor-specific structural realisations. The differences are quantitative (different step sizes, force ranges, processivity) rather than thermodynamically fundamental.

Population-averaged dynamics and the connection to physics §09 [Master]

In the large- $N$ limit (thousands to billions of cross-bridges in parallel), the discrete Markov cycle of individual heads is replaced by a continuous reaction-diffusion-style description of the cross-bridge density $n (x, t)$ along the actin sliding coordinate $x$ (Huxley 1957). The steady-state Hill force-velocity relation is one moment of this density.

The full mean-field dynamics, allowing the sarcomere length $L (t)$ and the velocity $v (t) = - \dot{L} (t)$ to vary in time, takes the form of a coupled system: the cross-bridge density $n (x, t)$ evolves by the rate-distortion PDE; the macroscopic force $F = F [n] = N k \int x n (x, t) d x$ provides the load on the sarcomere; the sarcomere geometry provides a constraint between $v$ and the rate of sarcomere shortening; and an external mechanical load completes the boundary condition.

This coupled system can be cast — under suitable smoothness assumptions and an effective elastic-energy functional for the sarcomere — in the form of a Hamiltonian system on a coarse-grained phase space whose generalised coordinates are the sarcomere lengths $L_{i} (t)$ of each half-sarcomere and whose conjugate momenta are the corresponding tensions $p_{i} = \partial L / \partial \dot{L}_{i}$ derived from the effective Lagrangian $L$ that reproduces the cross-bridge constitutive law. The Hamiltonian framework 09.04.02 pending gives the right description for energy-conservation arguments (mechanical work output as the Legendre-transform action), for normal-mode analysis of sarcomere oscillations (cardiac pacemaker auto-oscillations, insect flight-muscle resonance), and for the canonical-quantisation analogue (not biologically relevant, but mathematically present).

The reduction from microscopic Markov-cycle to macroscopic Hamiltonian-flow proceeds by: (i) coarse-graining over $N$ heads at each cross-section (law of large numbers), giving a hydrodynamic-like equation for the population; (ii) adiabatic elimination of the fast cross-bridge kinetics if the timescale separation $τ_{cycle} ≪ τ_{sarcomere}$ holds (it does at low-frequency contraction, marginal during fast tetanic contraction); (iii) closure of the resulting moment hierarchy at the level of cross-bridge tension. The dissipative residue of the underlying cycle (entropy production at the chemical level) appears as a friction term in the effective continuum dynamics — strictly speaking, the system is not Hamiltonian but Hamiltonian-plus-dissipation, and the closer one wants the conservative-flow framework the more one has to factor out the ATP-driven energy injection as an external work term.

This reduction is a textbook example of the molecular-to-continuum bridge that occupies a large part of biophysics: a system that is microscopically stochastic and chemical becomes macroscopically deterministic and mechanical, with each level having its own natural mathematical framework (Markov chains below, PDEs in the middle, Hamiltonian systems above). The level-matching is the substantive content of cross-bridge theory.

Cardiac and smooth muscle physiology [Master]

The actin-myosin core is conserved across striated and smooth muscle, but the regulation and the integrated physiology differ in ways that have major systems-physiology and clinical consequences.

Cardiac muscle. Cardiac myocytes are striated and use the same thin-filament regulatory machinery as skeletal muscle, with cardiac-specific isoforms (cTnI, cTnT, cTnC). Two cardiac-specific features dominate the physiology:

Calcium-induced calcium release (Fabiato 1985): a small trans-sarcolemmal $Ca^{2 +}$ entry through L-type channels during the cardiac action potential triggers a much larger $Ca^{2 +}$ release from the SR via the cardiac ryanodine receptor RyR2. The amplification factor is $\sim 10$ . This is a chemical coupling, whereas skeletal-muscle DHPR-RyR1 coupling is mechanical.
Length-dependent activation and the Frank-Starling mechanism: cardiac sarcomeres at longer length develop greater active tension at given $[Ca^{2 +}]$ , because the myofilament calcium sensitivity itself depends on sarcomere length. The mechanism involves changes in lattice spacing between thick and thin filaments, in tropomyosin-troponin azimuthal positioning, and in cooperativity of cross-bridge activation along the thin filament. The Frank-Starling mechanism is what allows the heart to match output to filling without extrinsic neural or hormonal regulation: a heart filled more (longer end-diastolic sarcomere length) pumps harder on the next beat. This is the cellular basis of one of the most important auto-regulatory mechanisms in physiology.

Smooth muscle. Smooth-muscle cells are spindle-shaped, lack sarcomeres, and have actin and myosin filaments organised in less regular networks anchored to dense bodies in the cytoplasm and to membrane plaques. Regulation is fundamentally different:

The regulatory step is on the thick filament: $Ca^{2 +}$ binds calmodulin in the cytoplasm; Ca-calmodulin activates myosin light-chain kinase (MLCK); MLCK phosphorylates the regulatory myosin light chain on serine-19; phosphorylation activates the myosin ATPase and allows cycling on actin. Thin-filament caldesmon and calponin provide additional regulation.
Slow contractions, low ATP cost. Smooth muscle myosin has a much slower cycle (heavy-chain isoforms with prolonged attached-state lifetime) — high force per unit ATP consumed, but slow shortening. The "latch state" further extends the attached-state lifetime at low calcium, allowing sustained tone (e.g., arterial smooth muscle maintaining vascular resistance) with minimal energetic cost.

The three muscle types thus exploit the same actin-myosin machinery for different physiological roles: skeletal for fast, voluntary, fatiguable contractions; cardiac for sustained rhythmic pumping with length-dependent auto-regulation; smooth for slow, sustained tone in hollow organs.

Connections [Master]

Hamilton's equations 09.04.02 pending provide the classical-mechanical framework for the coarse-grained dynamics of cross-bridge populations in the large- $N$ limit. The Hamiltonian-on-phase-space picture is the apex of the reduction from microscopic Markov cycle to macroscopic continuum mechanics described above.
ATP hydrolysis [15.14.NN, pending] is the chemical step that drives the cross-bridge cycle. The chem-side mechanism of phosphoanhydride bond cleavage at the myosin active site, the thermodynamics of the released phosphate, and the metalloenzyme chemistry of the catalytic site live in chem §15; the bio side cites these as the source of the chemical free-energy $Δ μ_{ATP}$ .
Cellular contractile proteins [17.03.NN, pending] — the cell-biology chapter on the actin and microtubule cytoskeleton — uses the muscle actin-myosin cycle as the canonical case study from which non-muscle myosin II (cytokinesis, cell migration, tissue tension) and the other myosin classes (V, VI, etc.) are conceptual extensions.
Action potential [17.09.02, pending] is the upstream excitatory event that initiates excitation-contraction coupling. The action potential's propagation along the sarcolemma and into the T-tubule system is the input signal to which calcium release responds.
Cardiovascular physiology [18.02.NN, pending] — particularly cardiac mechanics, the Frank-Starling mechanism, and cardiac pump function — is built on cardiac-muscle cross-bridge physiology as its molecular substrate.
Locomotion mechanics [18.04.NN, pending] — gait analysis, joint kinematics, and the mechanics of articulated skeletons — uses the actin-myosin cycle as its constitutive law for muscle force generation. The Hamiltonian-mechanical framework for articulated bodies cites both the physics-§09 mechanics foundation and this physiological-substrate unit.
Animal physiology — comparative [18.15.NN, pending] — the diversity of muscle physiology across taxa (insect flight muscle with stretch-activated oscillation, scallop catch muscle with paramyosin latch, asynchronous flight muscle in flies) — uses the canonical vertebrate skeletal cycle as a baseline against which comparative variation is described.
Stochastic thermodynamics [11.NN.NN, pending] in physics §11 — the modern framework for driven non-equilibrium cycles — has the muscle cross-bridge as its prototype biological example. Seifert's review treats actin-myosin in detail.
Mathematical modelling of muscle [02.NN.NN, pending] — the PDE Huxley 1957 model, the ODE Hill model, the Markov-chain cycle, and the modern Spudich-Goldman-Smith stochastic models — sit on math §02 ODE / PDE / probability foundations.
Phil-of-biology essay on agency [20.essays.07, proposed] — the actin-myosin cycle is a foundational test case for phil-of-bio's treatment of agency: a system that is purely mechanochemical at the molecular level (a Markov chain with chemical drive) generates apparently goal-directed organismal behaviour. The essay can take the cycle as its working example.
Cytoskeleton and contractile proteins 17.03.02 pending. The molecular-level actin-myosin contractility analysed in 17.03.02 pending — single-molecule force-velocity relationships, tight ATP coupling, ensemble force summation — is the direct prerequisite for the tissue-level treatment of sarcomere organisation, excitation-contraction coupling, and whole-muscle biomechanics developed here. The cellular cytoskeleton unit provides the molecular mechanism; this unit scales it to organ-level physiology.

Historical & philosophical context [Master]

The sliding-filament hypothesis was proposed simultaneously in 1954 by two independent groups: Andrew F. Huxley and Rolf Niedergerke at Cambridge ^{[Huxley-Niedergerke 1954]}, and Hugh E. Huxley (no relation) and Jean Hanson at MIT ^{[Huxley-Hanson 1954]}, in adjacent papers in Nature 173, May 22, 1954. Both groups had been studying X-ray diffraction and electron micrographs of striated muscle, and both arrived independently at the conclusion that contraction proceeded by relative sliding of two interdigitating filament systems whose lengths did not change. The convergence of evidence (X-ray spacing constancy of the A-band, electron-microscopic visualisation of overlapping filaments) was decisive within a few years.

The Hill force-velocity relation predates the sliding-filament model by sixteen years. A. V. Hill (Nobel 1922 for muscle heat measurements) had measured heat output and shortening velocity in tetanized frog sartorius muscle through the 1920s and 1930s, and published the hyperbolic force-velocity law in 1938 ^{[Hill 1938]} as an empirical fit. The relation was thermodynamic and phenomenological — no molecular interpretation was attached at the time, and Hill's own theory of "active points" anticipated cross-bridges only schematically. The molecular interpretation became possible with the cross-bridge picture in the late 1950s and was made quantitative by Huxley's 1957 cross-bridge kinetics model, which derived the Hill curve from a microscopic mechanism for the first time.

The cycle itself was elucidated by enzyme-kinetics work through the 1960s. Lymn and Taylor in 1971 ^{[Lymn-Taylor 1971]} integrated the kinetic data on myosin ATPase (Eisenberg, Trentham, Taylor labs) into the four-state cycle scheme that, with the rapid-hydrolysis-step insertion, became the standard five-state Lymn-Taylor cycle. The same year, A. F. Huxley and R. M. Simmons ^{[Huxley-Simmons 1971]} proposed the two-state attached-state model accommodating the T1/T2 rapid transients observed in length-step experiments on single fibres — this was the next refinement after Huxley 1957 and remains the textbook treatment of fast-transient cross-bridge mechanics.

The structural and single-molecule revolution of the 1990s — Rayment's 1993 myosin S1 X-ray structure ^{[Rayment 1993]} and the Finer-Simmons-Spudich 1994 laser-trap single-molecule measurements ^{[Finer-Simmons-Spudich 1994]} — closed the loop between kinetics, structure, and single-molecule mechanics. The structural and mechanical understanding of myosin in 2026 is far more detailed than it was in 1990, but the framework articulated by the 1954-1971 generation has not been overturned; subsequent work has refined the rate constants, identified the molecular conformational changes, and extended to other myosin isoforms and motor families, all within the basic sliding-filament cross-bridge cycle picture.

Two Nobel Prizes anchor the field: A. V. Hill in 1922 (for muscle heat measurements) and A. F. Huxley together with Hodgkin and Eccles in 1963 (for action potential and synaptic transmission; A. F. Huxley's muscle work was cited but not specifically the prize-eliciting result). Hugh Huxley, the other founder of sliding-filament theory, was not awarded the prize. The Rayment-Spudich generation of contributors has been honoured by the Lasker Award and various national academies but not yet by the Nobel.

Bibliography [Master]

Primary literature (cite when used; not all currently in reference/):

Huxley, H. E. & Hanson, J., "Changes in the cross-striations of muscle during contraction and stretch and their structural interpretation", Nature 173 (1954), 973–976. [Need to source — originator paper for sliding-filament hypothesis, MIT group.]
Huxley, A. F. & Niedergerke, R., "Structural changes in muscle during contraction; interference microscopy of living muscle fibres", Nature 173 (1954), 971–973. [Need to source — companion originator paper, Cambridge group.]
Hill, A. V., "The heat of shortening and the dynamic constants of muscle", Proc. Roy. Soc. London Ser. B 126 (1938), 136–195.
Huxley, A. F., "Muscle structure and theories of contraction", Prog. Biophys. Biophys. Chem. 7 (1957), 255–318. [Need to source — originator paper for rate-distortion cross-bridge model.]
Lymn, R. W. & Taylor, E. W., "Mechanism of adenosine triphosphate hydrolysis by actomyosin", Biochemistry 10 (1971), 4617–4624. [Need to source — originator paper for the four-state kinetic scheme.]
Huxley, A. F. & Simmons, R. M., "Proposed mechanism of force generation in striated muscle", Nature 233 (1971), 533–538. [Need to source — T1/T2 transient analysis, two-state attached model.]
Gordon, A. M., Huxley, A. F. & Julian, F. J., "The variation in isometric tension with sarcomere length in vertebrate muscle fibres", J. Physiol. 184 (1966), 170–192. [Need to source — quantitative length-tension data.]
Rayment, I., Rypniewski, W. R., Schmidt-Bäse, K., Smith, R., Tomchick, D. R., Benning, M. M., Winkelmann, D. A., Wesenberg, G. & Holden, H. M., "Three-dimensional structure of myosin subfragment-1: a molecular motor", Science 261 (1993), 50–58. Companion paper: Rayment, I. et al., Science 261 (1993), 58–65 (actin-myosin docking).
Finer, J. T., Simmons, R. M. & Spudich, J. A., "Single myosin molecule mechanics: piconewton forces and nanometre steps", Nature 368 (1994), 113–119.
Fabiato, A., "Time and calcium dependence of activation and inactivation of calcium-induced release of calcium from the sarcoplasmic reticulum of a skinned canine cardiac Purkinje cell", J. Gen. Physiol. 85 (1985), 247–289.
Seifert, U., "Stochastic thermodynamics, fluctuation theorems and molecular machines", Rep. Prog. Phys. 75 (2012), 126001. [Need to source — modern framework for non-equilibrium driven cycles.]

Textbook references:

Boron, W. F. & Boulpaep, E. L., Medical Physiology, 3rd ed. (Elsevier, 2017), Ch. 9.
Alberts, B., Johnson, A., Lewis, J., Morgan, D., Raff, M., Roberts, K. & Walter, P., Molecular Biology of the Cell, 6th ed. (Garland Science, 2014), Ch. 16.
Hill, R. W., Wyse, G. A. & Anderson, M., Animal Physiology, 4th ed. (Sinauer, 2016).
Squire, J. M., The Structural Basis of Muscular Contraction (Plenum, 1981).
Phillips, R., Kondev, J., Theriot, J. & Garcia, H., Physical Biology of the Cell, 2nd ed. (Garland, 2012), Ch. 16 (Molecular motors).

Wave 1 biology seed unit, produced 2026-05-18 by agent draft per docs/plans/BIOLOGY_PLAN.md §6 — second in the bio sequence after Hardy-Weinberg, before action potential. Cross-domain prereqs to chem §15 ATP biochem and bio §17 cellular contractile proteins are pending in the integrator backlog (proposed deps.json additions to be applied by integrator). Cross-domain hooks_out to physics §09 (Hamilton's equations, shipped), chem §15 (ATP hydrolysis, pending), bio §17 (cellular cytoskeleton, pending), bio §18.02 (cardiac mechanics, pending), and phil §20 (essay on agency) — all proposed until receiving-domain reviewers attest. Status remains draft pending Tyler's review and external physiologist reviewer per BIOLOGY_PLAN §7.

Prerequisites

09.04.02 pending
15.14.01 pending
17.03.02 pending

Tier anchors

beginner: Crash Course Anatomy & Physiology (muscle episodes 21–22); Campbell *Biology* intro chapters on muscle physiology; Khan Academy muscle modules
intermediate: Boron-Boulpaep *Medical Physiology* 3rd ed. Ch. 9 (Cellular Physiology of Skeletal, Cardiac, and Smooth Muscle); Alberts et al. *Molecular Biology of the Cell* 6th ed. Ch. 16 (The Cytoskeleton, §16.4 Molecular Motors); Campbell *Biology* 11th ed. Ch. 50 (Sensory and Motor Mechanisms)
master: Boron-Boulpaep advanced sections (Ch. 9 §III–V); Hill *Animal Physiology* 4th ed. (muscle and locomotion); Squire *Molecular Mechanism of Muscular Contraction* (Springer 1981) and Squire *The Structural Basis of Muscular Contraction* (Plenum 1981); primary literature — Huxley & Hanson 1954, Hill 1938, Lymn & Taylor 1971, Huxley & Simmons 1971, Rayment et al. 1993, Finer-Simmons-Spudich 1994

References

TODO_REF pending
Boron, W. F. & Boulpaep, E. L. — Medical Physiology, 3rd ed. (Elsevier, 2017) · Ch. 9 Cellular Physiology of Skeletal, Cardiac, and Smooth Muscle · see docs/catalogs/NEED_TO_SOURCE.md#bio-boron-boulpaep
TODO_REF pending
Alberts, B. et al. — Molecular Biology of the Cell, 6th ed. (Garland Science, 2014) · Ch. 16 The Cytoskeleton, §16.4 Molecular Motors; sarcomere and cross-bridge cycle · see docs/catalogs/NEED_TO_SOURCE.md#bio-alberts-mboc
TODO_REF pending
Hill, A. V. — The heat of shortening and the dynamic constants of muscle (Proc. Roy. Soc. B 126, 136–195, 1938) · Originator paper for the force-velocity relation in tetanized muscle · see docs/catalogs/NEED_TO_SOURCE.md#bio-hill-1938
TODO_REF pending
Huxley, H. E. & Hanson, J. — Changes in the cross-striations of muscle during contraction and stretch and their structural interpretation (Nature 173, 973–976, 1954) · Originator paper for the sliding-filament hypothesis; simultaneously with A. F. Huxley & Niedergerke 1954 same issue · see docs/catalogs/NEED_TO_SOURCE.md#bio-huxley-hanson-1954
TODO_REF pending
Huxley, A. F. — Muscle structure and theories of contraction (Prog. Biophys. Biophys. Chem. 7, 255–318, 1957) · Rate-distortion / two-state cross-bridge model — originator quantitative cross-bridge theory · see docs/catalogs/NEED_TO_SOURCE.md#bio-huxley-1957
TODO_REF pending
Lymn, R. W. & Taylor, E. W. — Mechanism of adenosine triphosphate hydrolysis by actomyosin (Biochemistry 10, 4617–4624, 1971) · Originator paper for the four-state Lymn-Taylor kinetic scheme of the cross-bridge cycle · see docs/catalogs/NEED_TO_SOURCE.md#bio-lymn-taylor-1971
TODO_REF pending
Huxley, A. F. & Simmons, R. M. — Proposed mechanism of force generation in striated muscle (Nature 233, 533–538, 1971) · T1/T2 transient response; elastic-element model for the power stroke · see docs/catalogs/NEED_TO_SOURCE.md#bio-huxley-simmons-1971
TODO_REF pending
Rayment, I. et al. — Three-dimensional structure of myosin subfragment-1: a molecular motor (Science 261, 50–58, 1993) · Chicken skeletal myosin S1 X-ray crystal structure; companion paper Rayment et al. Science 261, 58–65 on actin-myosin interface · see docs/catalogs/NEED_TO_SOURCE.md#bio-rayment-1993
TODO_REF pending
Finer, J. T., Simmons, R. M. & Spudich, J. A. — Single myosin molecule mechanics: piconewton forces and nanometre steps (Nature 368, 113–119, 1994) · Originator paper for single-molecule laser-trap measurements of myosin step size and force · see docs/catalogs/NEED_TO_SOURCE.md#bio-finer-simmons-spudich-1994
TODO_REF pending
Hill, A. V. — First and Last Experiments in Muscle Mechanics (Cambridge UP, 1970) · Retrospective synthesis of muscle thermodynamics, force-velocity, and length-tension · see docs/catalogs/NEED_TO_SOURCE.md#bio-hill-1970
tong
raw/pdfs/dynamics/four.pdf · Background reference for the Hamiltonian-flow framework underlying the population-averaged cross-bridge dynamics treated at master tier; readers focused on the biology may skip

Reviewer

Tyler (pending external physiologist reviewer per BIOLOGY_PLAN §7 — top recruitment priority alongside molecular biologist; LLM-augmented review per REVIEWER_PLAN until human physiologist is recruited)

Estimated time

beginner: 18m
intermediate: 40m
master: 65m