Sliding Filament Theory

The intuition: a muscle is a ratchet, not a sponge

Squeeze your bicep and it bulges — it feels like the muscle is balling up, like a sponge squeezing water out. That mental picture is wrong, and the sliding filament theory is the correction. Inside the muscle, the molecular machinery does not contract by shrinking. Instead, two sets of interleaved protein rods — thin actin filaments and thick myosin filaments — slide past one another like the two halves of an extending ladder collapsing back together. The rods keep their exact lengths the whole time. What shortens is the overlap region between them.

The pulling is done by hundreds of little arms — myosin heads — sticking out of each thick filament. Each head reaches over, grabs the neighbouring actin filament, and gives it a tug toward the center, then lets go and grabs again a little further along. Thousands of heads doing this asynchronously, like a crowd hauling a rope hand-over-hand, drag the actin inward. Burn one molecule of ATP per tug. Multiply by the tens of thousands of sarcomeres stacked end-to-end in a single muscle fiber, and those nanometer-scale slides sum into the centimeters of movement that flex your arm.

How the sliding works, step by step

The contractile unit is the sarcomere, the segment of a myofibril running from one Z disc to the next. Thin filaments anchor to the Z discs at each end and point inward; thick filaments float in the middle, anchored to a central M line. The myosin heads bridge the gap. Here is the full sequence, from electrical signal to physical shortening:

1. Excitation arrives. An action potential from a motor neuron crosses the neuromuscular junction, depolarizes the muscle fiber's membrane (the sarcolemma), and travels inward along the T-tubules — invaginations that carry the signal deep into the fiber.
2. Calcium is released. The T-tubule depolarization is sensed by dihydropyridine receptors, which mechanically open ryanodine receptors on the adjacent sarcoplasmic reticulum (the muscle's internal calcium store). Ca²⁺ floods the cytosol, jumping from about 0.1 µM at rest to roughly 10 µM.
3. The actin sites are unblocked. Calcium binds troponin C. Troponin changes shape and drags the long tropomyosin strand off the myosin-binding sites on actin. This is the on-switch — myosin can finally reach actin.
4. Cross-bridge forms. A cocked, energized myosin head (carrying ADP + Pᵢ from a previous ATP hydrolysis) binds the now-exposed actin, forming a cross-bridge.
5. Power stroke. The head releases its bound inorganic phosphate (Pᵢ), which triggers the lever arm to swing through about 5–10 nm, dragging the actin toward the M line. ADP is then released. This is the force-generating step.
6. Detachment. A fresh ATP binds the now-empty myosin head. This binding — not the hydrolysis — is what makes the head let go of actin, breaking the cross-bridge.
7. Re-cocking. Myosin's ATPase hydrolyzes the ATP to ADP + Pᵢ, and that energy snaps the head back into its high-energy cocked position, ready to bind further along the actin. Repeat from step 4.
8. Relaxation. When the signals stop, the SERCA pump uses ATP to vacuum Ca²⁺ back into the sarcoplasmic reticulum. Troponin lets go of calcium, tropomyosin re-covers the sites, cross-bridges can no longer form, and the muscle relaxes (passive elastic elements, like the giant protein titin, help restore resting length).

Crucially, the heads cycle asynchronously — at any instant only a small fraction are attached. A whole muscle pull is the smooth statistical average of billions of these jerky little strokes.

The molecular cast

• Actin (thin filament). A double helix of globular G-actin subunits, about 7 nm wide and roughly 1 µm long. Each G-actin has one myosin-binding site. Anchored to the Z disc.
• Myosin II (thick filament). About 300 myosin molecules bundled tail-to-tail, ~1.6 µm long and ~15 nm wide, with two-headed cross-bridges projecting outward in a helical array. Each head is both the motor and the ATPase.
• Tropomyosin. A long coiled-coil that lies in the groove of the actin helix, physically blocking the binding sites at rest.
• Troponin complex. Three subunits — TnC (binds calcium), TnI (inhibitory, holds tropomyosin in the blocking position), TnT (binds tropomyosin). The calcium sensor of striated muscle.
• Titin. The largest known protein (~3 MDa, ~1 µm long), it tethers each thick filament to the Z disc, acting as a molecular spring that provides passive elasticity and keeps the thick filament centered.
• Calcium (Ca²⁺). The trigger ion, stored in and released from the sarcoplasmic reticulum.
• ATP. The fuel — required for the power-stroke cycle (detachment + re-cocking) and for the SERCA pump that drives relaxation.

What the microscope shows: band changes

The genius of the 1954 papers was reading the sarcomere's striped banding under light and electron microscopy. The bands change in a very specific way during contraction, and that pattern is the fingerprint of sliding rather than shrinking:

Band / zone	What it contains	Resting width	During contraction
A band	Full length of thick (myosin) filament	~1.6 µm	Unchanged — the proof of sliding
I band	Thin (actin) filament only, no overlap	~0.8 µm	Narrows
H zone	Thick filament only, no overlap	~0.2 µm	Narrows / disappears
Z disc to Z disc (sarcomere)	The whole repeating unit	~2.3 µm	Shortens to ~1.5 µm
M line	Central anchor of thick filaments	—	Z discs pulled toward it

The logic is airtight: if the A band stays the same width while the whole sarcomere shortens, the thick filament cannot be folding or coiling. The only way to keep the A band constant while the Z discs close in is for the thin filaments to slide deeper into the A band, shrinking the I band and H zone at its expense.

The numbers: sizes, forces, speeds, energy

Quantity	Value	Note
Power stroke displacement	~5–10 nm	Per myosin head, per cycle (optical-trap measurement)
Force per myosin head	~1–5 pN	Single-molecule measurement
Heads per thick filament	~300	Helical array along the filament
Whole-muscle force	~20–35 N/cm²	Per cross-sectional area (specific tension)
Resting cytosolic Ca²⁺	~0.1 µM	Rises ~100× to ~10 µM on activation
Sarcomere length (rest → contracted)	~2.3 µm → ~1.5 µm	~30–35% shortening
Optimal length for max force	~2.0–2.2 µm	Peak overlap of heads with actin
ATP per cycle	1 molecule	Hydrolyzed to ADP + Pᵢ
Head cycling rate	~5 cycles/s	During active contraction
Duty ratio (myosin II)	~5%	Fraction of heads attached at any instant
Latency to peak Ca²⁺	~few ms	After the action potential

Length-tension: why overlap sets force

Because force comes from heads contacting actin, the amount of overlap between thin and thick filaments directly sets how much force a sarcomere can make — the length-tension relationship, first quantified by Gordon, Huxley, and Julian in 1966 on single frog muscle fibers.

• Optimal length (~2.0–2.2 µm): every myosin head in the cross-bridge region can reach an actin site. Maximum force.
• Overstretched (>3.6 µm): thin and thick filaments barely overlap, few heads can engage, and force falls toward zero. This is why a fully stretched muscle is weak.
• Overshortened (<2.0 µm): thin filaments from opposite ends collide in the middle and the thick filaments butt against the Z discs, both of which interfere with cross-bridging and reduce force.

This is not academic: the length-tension curve is why a strength athlete's joint angle matters, why the heart's Frank-Starling mechanism (more filling stretches the sarcomeres toward optimal length, so a fuller heart pumps harder) works, and why muscles in a cast lose sarcomeres to re-optimize their length for the immobilized position.

Where it shows up: organisms, disease, and the dead

• Rigor mortis. After death, ATP production stops. With no ATP to bind myosin and release it from actin, every engaged cross-bridge stays locked. The body stiffens 2–6 hours after death, peaks around 12 hours, and resolves only as the proteins themselves degrade. It is the sliding filament theory written on a corpse — and a direct demonstration that ATP's job is detachment, not the pull.
• The heart (cardiac muscle). Same sliding mechanism, same troponin switch, but its calcium is partly imported from outside and amplified by calcium-induced calcium release. Mutations in cardiac myosin (MYH7) and myosin-binding protein C (MYBPC3) cause hypertrophic cardiomyopathy, a leading cause of sudden cardiac death in young athletes.
• Smooth muscle. Blood vessels and gut walls slide actin and myosin too, but have no troponin or sarcomeres; calcium acts through calmodulin and myosin light-chain kinase. Their slow, sustained "latch" contraction at very low ATP cost is why an artery can hold tone for hours.
• Muscular dystrophies. Duchenne muscular dystrophy is a defect in dystrophin, the protein linking the contractile machinery to the cell membrane; the sliding still works, but the force is no longer transmitted safely to the outside, and fibers tear and die.
• Drugs that grab the cycle. Mavacamten (approved 2022) reduces myosin's ATPase activity to treat hypertrophic cardiomyopathy. Omecamtiv mecarbil does the opposite, increasing cardiac cross-bridge engagement in heart failure. Botulinum toxin blocks the acetylcholine that triggers the action potential upstream, paralyzing the muscle before sliding can begin.
• Insect flight muscle. Some insects beat their wings far faster than nerve impulses arrive by using "asynchronous" stretch-activated muscle, where the act of being stretched by an antagonist re-triggers the cross-bridge cycle — a remarkable variation on the same sliding hardware.

Common misconceptions

• "The filaments contract / shrink." No. Actin and myosin keep their lengths constant. Only their overlap changes, which is exactly why the A band width never changes.
• "ATP powers the pulling motion." Not directly. The power stroke is driven by the release of the already-bound Pᵢ and ADP. ATP is spent on detaching the head and re-cocking it. Rigor mortis (no ATP → permanent attachment) is the proof.
• "Calcium directly grabs the actin or the myosin." Calcium binds troponin, which moves tropomyosin off the binding sites. It is a switch on the thin filament, not a glue. (Smooth muscle is the exception — there calcium works via calmodulin on the myosin side.)
• "A bigger nerve signal makes one fiber pull harder per stroke." A single power stroke is roughly fixed in size. Whole-muscle force is graded by recruiting more motor units and by firing them faster so twitches summate (temporal summation, leading to tetanus), not by making individual strokes bigger.
• "All the myosin heads pull at once." Only about 5% are attached at any instant (the low duty ratio). Asynchronous cycling is what lets the filament slide smoothly instead of jamming.
• "Muscles actively push to extend." Muscles can only pull. Re-lengthening is passive — done by an antagonist muscle, gravity, or the elastic recoil of titin and connective tissue. There is no power stroke in reverse.