Molecular Motors and Cargo

A single kinesin molecule, carried on a bead and set on a microtubule in a dish, moves in jumps of 8 nanometres. At low load it spends one ATP per jump, and it takes about 100 jumps before it lets go.

Those numbers come from purified proteins in vitro, and they show what a motor protein is: an enzyme that turns ATP hydrolysis into steps along a track. Cells use motors to carry cargo, pull chromosomes apart and contract muscle. This lesson asks three things. Which motors walk on which tracks, and which way? How do you watch one molecule step? And how do millions of motors shorten a muscle when no filament gets shorter?

The idea: three families, two tracks, a direction per motor

Think of trains on one-way rails: each rail has a built-in direction, and each train runs one way. The rails here are the two polar polymers from the lesson on cytoskeletal dynamics. An actin filament has a barbed (plus) end and a pointed (minus) end. A microtubule has a fast-growing plus end, which in most animal cells points toward the cell edge, and a minus end, which usually sits near the centrosome.

• Myosins walk on actin. Most, including muscle myosin II and myosin V, move toward the barbed end. Purified myosin VI moves toward the pointed end.
• Kinesins walk on microtubules. Conventional kinesin (kinesin-1) moves toward the plus end, so it carries cargo outward. Not all do: ncd, a kinesin-related protein from fruit flies, moves toward the minus end in vitro.
• Dyneins walk on microtubules toward the minus end, so cytoplasmic dynein carries cargo inward. In a nerve cell it brings signals from the axon tip back to the cell body, as in the lesson on the retrograde signal and long-distance signaling in neurons.

So the family tells you the track but not always the direction. Direction belongs to each motor and has to be measured.

Step size also belongs to each motor, and to the conditions of the measurement. A bead carried by one kinesin advances 8 nm per step. Purified myosin V advances about 36 nm. Human dynein bound to dynactin and an adaptor, imaged without load at low ATP, took forward steps anywhere from 8 to 32 nm and often stepped backward or sideways.

Kinesin-1 and myosin V are processive: one molecule takes many steps in a row without letting go of the track. Each has two heads that alternate, hand over hand, so one head holds on while the other moves. For kinesin, the cargo moves 8 nm per step while each head travels about 16 nm, passing its partner; a dye on one head moved about 17 nm per step in a 2004 imaging study. Muscle myosin II works differently. Each head binds, pulls once and lets go, and force comes from many heads acting together.

What sets step size? In myosin V, the head is followed by a stiff neck, the lever arm, held straight by light chains. In a 2002 study, engineered myosin V with six light chains per neck stepped 36 nm, and a version with four stepped 24 nm. Shorten the lever and the stride shortens.

Watching one motor: gliding assays and optical traps

Kinesin was found by asking what moves organelles along nerve axons. In 1985 Ronald Vale, Thomas Reese and Michael Sheetz partly purified a protein from squid giant axons and optic lobes that moved microtubules, and named it kinesin. Scott Brady reported a candidate motor fraction from chick brain the same year.

The gliding assay turns the experiment upside down. Stick motors by their tails to a coverslip, add fluorescent filaments and ATP, and watch with a fluorescence microscope (microscopy lesson). The motors cannot move, so the filament does. Each motor walks toward its preferred end, which pushes the filament the other way, so a plus-end motor fixed to glass drives its microtubule minus end first. In 1986 Stephen Kron and James Spudich used this setup with fluorescent actin over myosin. Filaments moved only with ATP. Rabbit skeletal muscle myosin moved them at 3 to 4 µm per second, myosin from the amoeba Dictyostelium at 1 to 2.

Gliding measures many motors at once. To see one, use an optical trap: a focused laser that holds a small bead near its centre. A bead pulled off centre feels a restoring force, like a weight on a spring:

F = −κx, where κ is the trap stiffness and x is the bead’s distance from the trap centre

Coat beads so sparsely that most carry at most one motor, set one on a track, and record its position. In 1993 Karel Svoboda, Christoph Schmidt, Bruce Schnapp and Steven Block saw kinesin beads advance in 8-nm steps. In 1994 Svoboda and Block found that kinesin kept moving against loads up to 5 to 6 pN, slowing roughly in proportion to the load.

Worked problem: steps, force and efficiency

These data are made up, though in the range of real measurements. A bead carried by one kinesin walks along a microtubule in vitro, with plenty of ATP.

1. Step rate. At low load the bead moves 400 nm in 0.5 s, in 8-nm steps. Velocity = 400 / 0.5 = 800 nm/s. Steps per second = 800 / 8 = 100. Mark Schnitzer and Block concluded in 1997 that near zero load kinesin uses one ATP per 8-nm step, so this motor burns about 100 ATP per second.
2. Stall force. The trap stiffness is κ = 0.05 pN/nm. The motor walks until it stalls with the bead 120 nm from the trap centre. F = 0.05 × 120 = 6 pN, at the top of the 5 to 6 pN range Svoboda and Block measured.
3. Efficiency at stall. Work per step = force × distance = 6 pN × 8 nm = 48 pN·nm. Assume that hydrolysing one mole of ATP in the cell releases about 50 kJ (the true value depends on the concentrations of ATP, ADP and phosphate). Per molecule: 50,000 J / (6.022 × 10^23) = 8.3 × 10^-20 J. Since 1 pN·nm = 10^-21 J, that is 83 pN·nm. Efficiency = 48 / 83 ≈ 0.58, or about 58%.
4. A caution. The bead moves as far as the motor only if the link between them is stiff. A floppy tether stretches, and the bead under-reports the step.

A second caution: do not combine numbers from different experiments. In a 1994 trap study, one muscle myosin moved its actin filament about 11 nm at low load and pulled with 3 to 4 pN when held still. Those came from two loading conditions, so multiplying them does not give the work of one stroke.

Teams and adaptors: dynein-dynactin

Cytoplasmic dynein is built differently. Its motor is a ring of six AAA+ ATPase domains, with a long stalk that reaches the microtubule. In reconstitution experiments, purified human dynein becomes a fast, processive motor when it binds a second large complex, dynactin, together with an activating adaptor, a protein that also binds cargo (BICD2, BICDR1 and HOOK3 are examples).

The adaptor also sets how many motors ride together. In a 2018 study of human dynein with pig brain dynactin, cryo-electron microscopy showed that the adaptors BICDR1 and HOOK3 mostly recruited two dynein dimers, and those complexes produced more force and moved faster. Does the second dynein cause the extra speed? A 2019 study separated the two effects: adding a second dynein raised force but did not by itself set velocity, which depended on the adaptor and on how the closely packed dynein tails interact. In that system, a second dynein clearly bought pull; any effect on speed depended on the adaptor.

Teams can also pull against each other: when many kinesins drove one microtubule in a 2019 in vitro assay, about a third were slowing it. In cells, many cargoes carry both kinesin and dynein, and how a cargo settles on a direction is still being worked out.

Muscle: filaments slide, they do not shrink

A skeletal muscle fibre is striped. The repeating unit, the sarcomere, runs from one Z line (a protein disc) to the next. Thin filaments of actin anchor at each Z line and point inward. Thick filaments of myosin II sit in the middle. The A band spans the thick filaments; the I band is thin filament only; the H zone is the middle stretch of thick filament with no thin filament.

In 1954, in back-to-back papers in Nature, two pairs of researchers reported what happens to these bands as muscle shortens. Andrew Huxley and Rolf Niedergerke used interference microscopy of single living frog fibres. Hugh Huxley and Jean Hanson used phase-contrast microscopy of isolated myofibrils. (The two Huxleys were not related.) In both, the A band kept its width while the I band narrowed. If the A band stays the same, the thick filaments are not shortening. The thin filaments slide in between them, the H zone narrows, and the Z lines are drawn together.

Sliding predicts a length-tension curve. In 1966 A. M. Gordon, Andrew Huxley and F. J. Julian held single frog fibres at fixed sarcomere lengths and measured tension during contraction. Tension peaked on a plateau between 2.05 and 2.2 µm. Stretched further, tension fell roughly in line with the lost overlap between thin filaments and myosin heads. At shorter lengths it fell too, as the thin filaments crowd past one another.

What makes the filaments slide is the cross-bridge cycle. A myosin head carrying ADP and phosphate binds actin. Phosphate release goes with the power stroke, which pulls the thin filament toward the middle of the sarcomere. ADP leaves, and the head stays tightly bound. ATP binding releases it from actin, and hydrolysis cocks the head for the next cycle. With no ATP, heads stay locked to actin, which is why muscles stiffen after death (rigor mortis). In living muscle, calcium decides when heads may bind, through troponin and tropomyosin on the thin filament (calcium signaling covers that switch).

Design problem

You purify a new microtubule motor from a cell extract and must find which way it walks. Design the experiment, the controls, and the result that would tell you it is a plus-end motor.