User:Kinkreet/Protein Science/Molecular Motors

Motors are devices which converts a source of energy to useful (directed) mechanical force/action. Most of the movements in life are powered by molecular motors. These motors transport cargo (myosin, kinesin and dynein), organelle movement (dynein), involved in cell locomotion (flagella movement using dynein), energy generation (F0F1-ATPase), protein-nucleic acid dissassembly (DNA and RNA helicases), disassembly of protein aggregates or complexes (N-ethylmaleimide sensitive fusion protein, NSF, and p97, an AAA family ATPase), cell division, muscle contraction (myosin II), and have other functions.

Protein motors are involved in. Kinesin, myosin, dynein, linear DNA helicase, ring DNA helicase and F0F1-ATPase are amongst the most well-understood protein motors.

On the molecular level, the majority of the energy source is provided chemically by ATP. There are three main types of motors: those that move along a track (linear; such as kinesin, myosin and dynein), those that rotates (rotary), but the most efficient motors couples linear and rotary movement together.

Optical Trapping
Optical trapping, also known as optical tweezers, is a method to move or hold microscopic dielectric (electrically insulating but can be polarized in an electric field) objects using highly focused laser beams. It is based on the principle that bending a ray of light involves transfer of momentum. It can work on particles as small as an atom. In biology, it is used to apply forces in the pN range to measure displacements in the 10-100 nm range.

It has been used to trap dielectric spheres, viruses, bacteria, living cells, organelles, small metal particles and DNA strands.

The laser beam is focused to a single point (called the beam waist) on the specimen plane; the intensity of the beam has a Guassian distribution, which means it is brightest in the centre compared to the edges. When light hits the bead, the direction of propagation is changed dependent on the reflective and refractive properties of the bead. This change in direction results in a change in momentum.

The forces generated from this change of momentum is split into two components: Fscattering, the scattering force, pointing in the direction of the incident light; and Fgradient, the gradient force, arising from the gradient of the Gaussian intensity profile and pointing in x-y plane towards the center of the beam. It is Fgradient which forces the bead to move towards, or be maintained at, the centre of the beam.

Whenever the dielectric particle is displaced, it will try to move back into the trap. If we do an experiment where the path of the kinesin motor crosses the beam waist, when the cargo (a glass or polystyrene bead) passes the beam waist and is moving away, a force will be applied to the bead pulling it back towards the waist. The force is proportional to the displacement, given that displacement is small; during these conditions, the optical trap can act as a spring, following Hooke's law.

The displacement can be calculated after the refracted and reflected rays are measured by a quadrant photodiode. From the displacement, we can elucidate the forces on the bead at each point if we consider the trap to act as a spring, and thus the minimum forces required by the motor to continue walking.

From these data, we can elucidate whether the motor takes individual steps, and if so its step size, and the force of the motor.

Structure
Linear motors include kinesins, myosins, dyneins, PcrA DNA Helicase etc. All linear motors apart from DNA helicase have the same general structure, containing a dimer of motor domains, which is connected by a long linker (a.k.a. arm or light/heavy chains), to which a stalk that has a cargo-binding domain extends from. The motor domain is connected to the long linker via the neck linker in kinesin, and the converter domain in myosin; their similarities in structure led to the hypothesis that the two proteins share common ancestors.

Function
Traditionally, it was thought that the three motors have distinct and separate functions: myosin is involved in contraction, kinesin in organelle transport and dynein with ciliary beating. It is now known some of their functions overlap, for example, all three motors are required for organelle transport and cell division. New uses, such as signalling and signal transduction and RNA localization, have also been realized.

Motors can be localized on membranes by their tail regions. Myosins with a basic tail can bind to acidic phospholipids; kinesins of the Unc104/KIF1 family have a pleckstrin domain which binds PI(4,5)P2. Lipid rafts which is high on acidic phospholipids, or PI(4,5)P2 will see localization of their respective motors. The tail region is also used to bind cargo, mostly organelles and vesicles, which are all lipid-based. This led to the hypothesis that the motors latch onto their cargo by binding to integral membrane proteins on the cargo. This binding can be direct (using a component of the motor itself), for example the tail domain of myosin I with acidic phospholipids or cytoplasmic dynein binding to rhodopsin via Tctex-1 light chain; or indirect (via a linker protein/complex), for example, dynein can bind to an integral membrane via a dynactin complex and spectrin, and myosin V can bind to rab27a via melanopholin. Kinesin light chains can bind to amyloid precursor protein, which enables the vesicle's transport along the axon. Failure to transport efficiently leads to formation of amyloid plaques, which causes Alzheimer's disease.

Mechanism
Kinesin, myosin, dynein amplify small conformational changes to allow them to take nanometer steps along protein tracks. Myosins have actin filaments as tracks, dyneins and kinesins move along microtubules. These motors work on the basic principle that chemical energy from ATP hydrolysis is converted to mechanical energy in the form of a small conformational change of a globular motor domain; the small conformational change is then amplified to various degrees using accessory structural motifs. This allows a small conformational change to give rise to large movements.

However, a large conformational change in a monomeric protein will not generate translational movement, for that to happen it must dimerize. Domains outside of the motor domain is responsible for dimerization.

Detailed
When ATP is hydrolysed, the γ-phosphate leaves the binding pocket, which leaves a space 0.5nm in diameter. This spaces causes the conformational change for the sequences flanking the ATP-binding pocket. This conformational change is coordinated so hydrolysis causes the track-binding site to change conformation to one which is able to bind the track.

The conformational change also affects a set of carbonxy-terminal proteins (α-helical stem in myosin and neck linker in kinesin) which acts as levers that amplifies the small conformational change into a more pronounced movement, although the force of the movement is the same. The set of proteins for this mechanical amplification differs between kinesins and myosins; in both myosins and kinesins, a switch II helix is essential for this relay. The relayed signal for ATP hydrolysis causes the converter or neck linker to become undocked from the core, and free.

Generally, the longer this lever, or stalk, the larger the steps taken by these motor proteins; myosin VI have a step size of 36nm, the shortest distance between two heads bound to F-actin is 24nm, and so the lever arm must be long enough to bridge that gap. Myosins have an α-helix stabilized by light chains, and can swing up to 70°; kinesins have a flexible neck linker that relocates.

The series of amplification allows the small conformational changed caused by the absence of a γ-phosphate from the hydrolyzed ATP, is able to be amplified to as much as a 36nm step (in myosin V).

Using microdevices, the force generated by myosins, kinesins and dynein is determined to be in the range 1-10 pN.

Variety
Within the myosin, kinesin families, a huge variety exists: some exists as monomers, dimers, trimers or tetramers, move in opposite directions, are processive or non-processive. The importance of each member of the family also differs; yeast have 6 kinesins, 5 myosins and 1 dynein, whereas mammals have over 40 kinesins, 40 myosins and more than a dozen dynein. The actual number of variations can 'easily' triple due to post-translational modifications and a variety of accessory proteins. Despite this variation, the initial conformational change generated from ATP hydrolysis for all motor proteins follow a similar mechanism. This might be due to the fact that the motor domains (where ATP hydrolysis occurs), is similar between the proteins. For example, the motor domain of myosins and kinesins share many structural similarities (elucidated using crystal structures): the structure surrounding the ATP-binding pocket is almost identical, although the sequence homology is only observed for a few residues. (Note that the ATP-binding site is similar to that of the G-proteins' GTP-binding site, suggesting that kinesins, myosins and G-proteins all have a not-too-distant common ancestor)

Processivity
There are some motors which are processive, meaning their intended function requires them to stay on the track without detaching for as long as possible, and there are some motors which makes lose contacts with the track and usually detach after one cycle. This variety is suited to the different motor's cellular function. One of the functions of conventional kinesin is to transport cargo along microtubules, therefore, it is progressive.

Kinesins are dimers that have ATP-hydrolysis cycles in opposite phase to each other. When the 'front' head is attached tightly onto the microtubule, the 'back' head swings towards its new binding site, in a 'hand-over-hand' fashion over the 'front' head. The role of each head is now reversed and the process recurs.

Myosin V is a dimer where the two heads are co-operative. On its neck, it has 6 light-chain-binding sites which allows myosin V to have a rigid and long (~36nm) stalk. The length of its stride coincides with the distance between pits on the actin filaments. This power stroke is made of two components - the energy from ATP hydrolysis provides a working stroke of ~25nm, and diffusion is used to travel the remaining 11nm. This is also observed for myosin VI. The working stroke provides a bias of the directionality of the motor, while the diffusion completes the stroke.

Muscle myosin II is a dimer where the two heads are not co-operative, and thus the myosin do not stay attached to actin for long; the time it makes solid contact with actin is less than 10% of the whole cycle. This non-processivity suits the function of myosin II, which is to provide force for muscle contraction; if it remains attached to actin, it will slow down the whole system.

Processive motors usually work as a stand-alone unit, whereas non-processive motors usually works as a group, with each member contributing a bit of the whole movement.

Processive motors are usually dimeric and require co-operative actions, but there are also progressive which are monomeric: KIF1A kinesin, monomeric class IXb myosin and monomeric inner arm dynein; but their mechanism is fundamentally different. For example, monomeric KIF1A step back and forth, but with a general direction towards the plus end of the microtubule. There is a positive K loop which interacts with the negative C-terminus of tubulin. The loop tethers the monomer onto the microtubule, while the power stroke pushes it towards the plus end. However, this K-loop tethering is only a component of the whole mechanism, as there are members of the KIF1A family containing the K-loop and is non-processive.

Directionality
Most kinesins and myosins move towards the plus end of microtubules and actin, respectively. However, minus end-directed kinesin-like proteins and a minus-end-directed myosin have also been observed. The kinesin-14 family, including the Drosophila ncd, budding yeast KAR3, and Arabidopsis ATK5, is one example. Most motors which move in this uncommon direction usually have its motor domain at the C-terminus rather than the N-terminus.

The determining factor in directionality of kinesins often lies with the neck of the motor. Chimeric motor proteins studies showed that when the motor domain of opposite-directed motor proteins, the resulting directionality depends only on the neck. Furthermore, a single-amino acid point mutation in the ncd neck, or the contact between the neck and the motor domain, causes the resulting motor to move with wild-type velocities, but without a distinct direction.

Myosin VI moves towards the minus end of actin; its reversed directionality with other myosins is attributed to a 53 amino acids insertion in the converter domain , which rotates its lever arm to point in the opposite direction to the conventional myosin, which changes the direction of the lever-arm swing by ~120°, and it flicks back in the opposite direction for ~180°; thus myosin VI moves towards the minus end. The motor domain is conserved and not reversed in direction.

From studying the crystal structure of myosin II, it has been shown that small conformational change in the motor domain is transmitted to the lever-arm through a converter domain. The lever arm consists of a light-chain-binding helix and associated light chains, which amplifies the small conformational change into a large movement. The mechanism contributing to the reversal of directionality of kinesins and myosins is different: kinesin's depends on the neck region, whereas myosins depend on the converter region.

Kinesin
Moves along randomly-orientated actin filaments The heads in kinesin is much closer together, because instead of a long lever arm, it is held together by a short polypeptide chain. This results in smaller steps. The smaller steps may mean the motor is less likely to fall off, as less of the step is attributed to diffusion, and also because the proportion of time where both heads are latched on to the track is higher. Their close proximity also means that conformational change in one head can be communicated to the neck linker of the second head, allowing the two heads to act cooperatively.

In contrast to myosin, when ATP is bound, it increases the binding surface and thus bind tighter to microtubule filaments (for myosin, nucleotide-free state provided the tightest binding).

With the forward head and neck linker docked, the trailing head (ADP-bound and have a free, flexible linker) diffuses in the general direction of the plus end, where it will dock. With both heads bound, the now-back head hydrolyzes its ATP to ADP and Pi; Pi dissociates and free the trailing head. The front head (newly docked) swaps its ADP for ATP and binds tightly to the microtubule; this binding is communicated to the neck linker of the now-trailing head, which causes it to reposition itself forward of the front head, dragging along the trailing head. ATP of the back head is hydrolysed and become free and the cycle repeats.

The two heads never exists at the same nucleotide state. The step size of kinesin is ~8nm, and a processivity of ~100 steps, at a speed of ~40 steps per second. The force of the motor is ~6pN.

Single-molecule studies
A fusion protein where the tail of kinesin is fused with gelsolin, an actin-binding protein. Gelsolin can bind to rhodamine-phalloidin-labelled actin, which is fluorescent and can be viewed under the microscope in real time to follow the movement of kinesin. It was found that the speed of kinesin is ~300nm per second, and have a processivity of 400-600nm.

This method have an advantage over GFP-kinesin, which is prone to photobleaching, and to optical trapping, which do not register the steps if the motor has no, or low load.

Function Appearence/Structure Step size and fashion, energy required (steps per ATP cycle) Directionality Mechanism - how is chemical energy converted to mechanical energy? what is the conformational change? Force Efficiency

Myosin
Myosin moves along radially orientated microtubule filaments towards the plus-end (usually towards the cell periphery). It has a step size of ~36nm, a processivity of ~40 steps, and a force of 1-5pN.

Nucleotide-binding motor domain
ATP binds to the nucleotide binding site. Switch I and II is close to the nucleotide binding site and propagate the binding signal, to the track(actin)-binding site, and also to the converter domain, which is linked to the lever arms, which consists of an alpha-helix with 6 light-chain binding sites, and 6 light chains. The crystal structure of the nucleotide-bound motor domain was elucidated using MgADP.BeFx, an analog of ATP which does not get hydrolysed after being bound.

Actin-binding
The track binding site is the cleft between the upper subdomain and the lower subdomain; when there are no nucleotide bound, the cleft is closed. This means if it has bound to actin, it will bind tightly, otherwise it would not bind at all. There is also an extended loop which associates with F-actin for a tighter binding.

After ATP binding and hydrolysis, but before ADP or Pi is dissociated (i.e. at the transition state), the loop that contributed to the actin interaction, is folded back, and this reduces the strength of binding to actin.

Swing
There are two relay helices which connects Switch I and II to the converter domain. The converter domain connects the motor domain to the long arm which links to the cargo binding domain.

When ATP binds, the converter domain shifts at a slight angle, but as soon as ATP is hydrolysed, the shift becomes the most significant. Because the lever arm is attached to the converter, a shift in conformation of the converter will mean a shift of position of the lever arm, leading to a 'power stroke'.

Myosin V
Myosin V is a processive (carries on for a long time) motor which steps along microtubules at 36nm per step. It is involved in the transportation of cargo, polarized cell growth, membrane trafficking.

Nucleotide-free myosin binds tightly to F-actin. Upon ATP binding, this interaction is weakened and myosin dissociates with actin (the dimer is held in place because the other myosin is bound to actin). After dissociation, ATP is hydrolysed and this causes the lever arm to perform a power stroke 'forward', and aided by diffusion to associate and bind to a 'pit' on the F-actin. Pi is dissociated quickly and this leads to both the heads being bound to F-actin. ADP of the other head is then dissociated slowly (at a rate of ~12 per second) to return it to the original state, where the head binds tightly.

Myosin II
Myosin II is similar in terms of ATP hydrolysis and conformational change, but it does not move forward or walk; instead it drags the thin filaments (F-actin).

The myosin head hydrolyzes ATP and swings 'forward', upon Pi dissociation, it binds to actin by forming crossbridges. ADP dissociation leads to the power stroke, where the head moves 'back' towards the centre of the sarcomere. ATP then binds and weakens interaction with actin, and so myosin II will dissociates.

The two heads of the dimer are not co-operative, and thus the myosin do not stay attached to actin for long; the time it makes solid contact with actin is less than 10% of the whole cycle. This non-processivity suits the function of myosin II, which is to provide force for muscle contraction; if it remains attached to actin, it will slow down the whole system.

The nucleotide cycle in both myosin V and II are similar, with the major difference being there is no power stroke in V, because the motor is not anchored and the energy generated is used to translocate itself in relation to the F-actin. There is a power stroke for myosin II because myosin II is anchored and cannot move. Efficiency

Dynein
Moves along microtubules towards the minus-end (usually towards the nucleus)

The dynein motor is much bigger and more complex than both myosin and kinesin; so until recently, the dynein motor lacks high-resolution and thus its mechanism is less well understood. But from its sequence homology, it can be said with confidence that the mechanism of the dynein motor is fundamentally different to that of both myosins and kinesins. The motor domain of dynein is made up of 7 subunits, consisting of 6 AAA ATPase modules forming a ring, plus one linker domain which spans across the ring. Of the 6 ATPases, only 1 is active. ATP-dependent conformational change is communicated from the ring to a stalk which has the microtubule-binding site at its tip; at the opposite end is a stem which contains the dimerization site and cargo-binding site. The swing of the stalk leads to ~8nm steps, but occasionally can have larger step size of 15-24nm steps. The ATP-binding site is between AAA1 and AAA2. When nucleotide-free, the linker domain sits across the ring and infringes on the ATP-binding site. When ATP binds, the linker is displaced and bend, which changes the conformation of the stalk and affect microtubule binding. Therefore, ATP-binding induces the release of microtubule. After release, the stem undergoes a priming stroke, moving the head towards the minus-end; at the end of the stroke, the head is primed and binds to microtubule. Upon hydrolysis of ATP and dissociation of ADP and Pi, the power stroke occurs. The motor domain returns to the nucleotide-free state, which is where the head binds the tightest.

Function Appearence/Structure

Dyneins belong to the AAA proteins, and can be classfied into two subfamilies: cytoplasmic dynein and axonemal dyneins. Cytoplasmic dynein aid in orientating the cell spindle during mitosis, nuclear migration and neuronal transport. Axonemal dynein are immobilized, and acts against each other to create the bending motions in cilia and flagella.

Step size and fashion Directionality Mechanism - how is chemical energy converted to mechanical energy? what is the conformational change? Force Efficiency

Helicases and translocases
Helicases are translocases are enzymes that use energy from ATP hydrolysis to separate or move nucleic acids. There are 5' → 3' helicases and 3' → 5' helicases.

Different mechanisms for helicase action have been proposed, the two main proponents are rolling vs. inchworm mechanism. The rolling mechanism requires a dimer, the inchworm hypothesis involves just a monomer.

PcrA is a monomeric helicase that contains an ATPase domain and two subdomains. It translocate using the inchworm mechanism and uses ssDNA as track. There are four nucleotide binding sites on PcrA - A, B, C, and D - with A binding the 3'-most nucleotide, and D binding the 5'-most two nucleotides. After PcrA binds, ATP is hydrolysed so Phe64 (which lines site B) pushes nucleotide at site B to A, nucleotides from site C then fills the empty gap left in B, nucleotides from D fills C and the effect passes along. This results in the translocation of the PcrA helicase along the ssDNA with a step size of 1 nucleotide.

By measuring the concentration of inorganic phosphate, released after ATP is hydrolysed, we can calculate the rate of ATP hydrolysis and thus its speed (because we already know its step length and energy requirements). From this we calculated the speed to be 55 bases per second.

F0F1-ATPase
F0F1-ATPase synthesizes ATP from ADP + P1 using a built-up proton gradient (or salt gradient) as a source of energy. The reverse is also possible, where it hydrolyzes ATP to pump protons across the membrane to build up a proton gradient.

The F0F1-ATPase is found on the inner membranes of bacterial entities, such as the inner membrane of mitochondria, the thylakoid membranes of chloroplasts, and bacterial plasmid membranes. F0F1-ATPase is made up of a F0 subunit, a F1 subunit, as well as other subunits (γ, ε, α, β, δ).

The F0 subunit is made up of 10-14 c subunits making up the C ring. The protons uses the C ring as a turbine so as it moves down its gradient, it rotates the C ring.

The F1 subunit exists outside the membrane. It itself is made up of 6 subunits - 3 α separated by 3 β - forming a ring. The α and β subunits are homologous, but only the β is catalytic, the α acts only as structural units.

The different β units of F1 exists in different nucleotide and conformational states. It goes from loose, where ADP and Pi can bind, to Tight, where ATP is synthesized, to open, where ATP is released. Therefore, each β subunit undergoes the nucleotide change of T → O → L → T → O → L. These different nucleotide states are determined by the interaction it has with a central γ subunit. As the γ subunit rotates, driven indirectly by the proton gradient, it induces conformational change in each of the subunits.

From single-molecule studies, using fluorescently-labelled actin to visualize rotation, it is found that rotation occurs in 120° steps, with a torque of 40pN. The efficiency of the enzyme is almost 100%.

Glu65 of the c subunit interacts with Arg227 of the ε subunit between the a and c subunits. When proton or sodium ions move down its gradient, it will meet Glu65 of the c subunit; the positive charge will neutralize Glu65, breaking its interaction with Arg227, and so Glu65 will move away, making way for Glu65 from another c subunit to bind. Because the central γ stalk is connected to the c ring, rotation of the c ring causes rotation of the stalk.

T7 gene 4 helicase
T7 gene 4 helicase is a 5' → 3' ring helicase. The ring consists of a homohexamer, each with a nucleotide binding site, but only 4 have been found to bind to ADPNP at any one time, two sites opposite each other are empty.

This is almost the reverse to the F0F1-ATPase; here ATP binds to the empty site, gets hydrolyzed and then releases ADP and Pi. The nucleotide of the six sites are synchronized, with opposite sites having the same nucleotide state at any one time. The nucleotide state corresponds to a conformation of an inside loop region of ~7Å. The synchronized conformational change of the six subunits are coupled to step-wise migration of DNA, must like a bolt (the ring helicase) climbing up a screw (the DNA).

E1 helicase
The E1 helicase is also a homohexameric ring DNA helicase, but instead of having two pairs of synchronized nucleotide states, it works in more of a rotary manner, where the nucleotide state of the six subunits go in the order of Free → ATP → ATP → ATP → ADP → ADP → Free → ATP...

Rho ATPase
Rho is a helicase which work on both RNA and DNA, and is involved in Rho-dependent transcription termination, which is used for ～50% of all mRNA transcription.

Rho consists of six identical subunits, organised into a slanted ring, or a skewed cup, with the flat side of each subunit creating an angle of 75° to a plane perpendicular to the ring axis ; when it needs to bind to the RNA nascent chain, the RNA-binding domains rearranges so as to open a gap in the ring, the gap is large enough for ssRNA to slide through. The structure of this split ring is determined by Skordalakes and Berger in 2003 by crystallography of two Rho proteins, one bound with a 15-mer A,C-rich ssDNA, and the other with an 8-mer RNA of (rC-rU)4, both together with a non-hydrolysable ATP analogue, AMPPNP. The first 125 amino acid residues (near the N-terminus) of each subunit forms a separate domain to the rest of the protein (419AA in length), and can bind to C-rich single-stranded nucleic acids, and thus is termed the RNA binding domain, Each Rho subunit has one RNA-binding domain consisting of two RNA-binding sites, one at the surface of the hexameric ring making up the primary site, and one towards the centre hole of the ring making the secondary site . The primary site is not specific for RNA, because it can bind to ssDNA also. The secondary site is where specific interactions occur, namely the activation of ATP hydrolysis and helicase function, both are specific to RNA.

The structure of the RNA binding domain consists of a three-helix bundle followed by a 5-stranded β-barrel with an OB fold; this structure is common in single-stranded nucleic acid binding proteins. Each RNA-binding site is thought to bind to two nucleotides at a time; and the RNA, in the crystal structure obtained by Skordalakes, binds to all six of the subunits. Since the minimal distance between the RNA-binding sites of adjacent subunits is 35Å, at least 60 nucleotides from a single RNA can be bound to the hexameric ring, provided it follows a zigzag pattern from one subunit to the next. This is consistent with the finding that at least 60 residues are required for Rho to cause termination.

In the secondary site, peptide loops (called Q-loops) extend into the hole in the ring, and forms the narrowest opening of the ring, and also the bottom of the skewed cup in the analogy.

The remaining 294 residues (at the C-terminal end) make up the ATP-binding and hydrolysis domain. This domain resembles part of the α and β subunits in F1 ATPase, and so this is used to model the ATP-binding domain. It is presumed that in the notched, or loading form, the ATPase is inactive because its binding site do not have the correct conformation; but when the Q-loops interact with the RNA, it changes the binding site’s conformation so it is able to bind ATP. It also closes the hexameric ring fully and the ATPase activity would cause rho to move towards the RNAP in a 5’ → 3’ direction through the centre of the ring, in order to terminate transcription. This hypothesis explains how rho only moves down the RNA once the RNA is loaded onto rho.

The ATPase domains change nucleotide states from T → T → T → T → D → E. The conformational change is also synchronized so the ring helicase moves up the nucleic acid like a bolt moving up a screw. However, unlike E1 helicase, the nucleotide changes occurs in the opposite direction in the ring (or that the hydrolysis cycle works in the opposite direction), and so instead of moving in a 5' → 3' direction, it moves in the opposite direction, and thus be able to move back to the polymerase and stop transcription.

In bacteria, translation and transcription is often coupled; this means that the ribosome might bind to the RNA (provided a translational initiation site is available) and start translating before rho has a chance to bind to terminate transcription. This is good because we do not want transcription to terminate prematurely. The ribosome will translate the transcript at the same pace as RNAP translocation, until it reaches the stop codon where the ribosome will stop. RNAP however will continue for a further 70 nucleotides before it reaches its termination sequence. By this time, there would be about 90 nucleotides of continuous ssRNA for rho to bind to. Of these 90 nucleotides, 8 would be protected by the ribosome, 60 would be taken up by the primary site of rho, and 12 would be required to pass the ssRNA through the Q-loop, finally 14 would be buried within the RNAP.

Studies have shown that a protein called NusG is required when rho and RNAP meet, it forms a bridge between the two and help rho to load the nascent RNA onto itself.

Rho do not load onto transcripts that are not translated, such as tRNA or rRNA. It does not load onto tRNA because it forms a secondary structure as soon as it can, so there is no ssRNA for rho to bind to; it does not load onto rRNA because rRNA transcript signals to the RNAP at the beginning of transcription to produce a complex that keeps rho away from the transcript

AAA ATPases
The AAA ATPases (ATPases Associated with diverse cellular Activities) is a very large family of ATPases. They are characterized by a conserved domain consisting of ~230 residues, which includes the Walker A and Walker B motifs (named after John Walker). It belongs to a large family of AAA proteins, characterized by ring-shaped P-loop NTPases. This ring shaped is formed from oligomers of monomeric AAA ATPases, typically being a hexamer. AAA ATPase couple the hydrolysis of ATPase with a conformational change which exerts force on some macromolecules; it is thus involved in DNA replication, protein unfolding, translocation and degradation (ClpXP and proteasome), disassembly of protein aggregates and complexes (p97), membrane fusion, microtubule severing, peroxisome biogenesis, signal transduction and activation of transcription (PspF).

The Walker A motif is also called the P-loop, for phosphate-binding loop; it has a conserved sequence of GXXXXGK(T/S). The lysine (K) in that sequence is essential for nucleotide binding. This motif is found commonly in nucleotide-binding proteins, but also in alpha and beta subunits of ATP synthase, myosin, transducin, helicases, kinases and RecA. The Walker B motif is defined originally by Walker to be a conserved sequence of (R/K)XXXXGXXXXLhhhhD, but later Hanson refined this sequence to hhhhDE.

The two amine groups of arginine stabilizes two oxygens at the terminal γ-phosphate. A magnesium also coordinates the β- and γ-phosphates. The hydroxide group of serine or threonine of the Walker A loop stabilizes this magnesium ion along with the carboxylate group of aspartate of the Walker B loop. The carboxylate of the glutamate hydrogen bond with a hydrogen of a water molecule, holding it in place so that the oxygen (from the water), can nucleophilically attack the γ-phosphate, causing ATP hydrolysis. Once the Pi dissociates, this leaves a ~0.5Å gap which is the source of the conformational change.

ClpXP Protease
Interruption of translation in E. coli can lead to an addition of a 11 residue sequence to the C-terminus by SsrA RNA (also called 10Sa RNA and tmRNA); this marks the protein for degradation. Proteolysis is carried out by the ClpXP protease.

The ClpXP (Caseinolytic protease XP) is a complex of many proteins, split into two main subunits of ClpX and ClpP. ClpX uses ATP hydrolysis to unfold native proteins and translocate polypeptides into the ClpP chamber, where proteolysis occurs.

Crystal structure shows ClpX is a homohexamer arranged in a symmetrical ring, and have 6 nucleotide binding sites. In the absence of substrate (the protein to be degraded) 4 nucleotides are bound while 2 remains empty; in the presence of substrate, four sites are bound with ATP, 1 with ADP and one is empty.

Proteasome
The proteasome is an ATP-dependent, multisubunit protease which is used to degrade ubiquinated intracellular proteins, both pathogenic and self, to create antigens. First, the E1 ubiquitin-activating enzymes uses ATP hydrolysis (to AMP) to bind to an ubiquitin protein. This ubiquitin is then passed on to E2, the carrier which complexes with E3 ubiquitin ligase to add the ubiquitin to the substrate. Some of the polyubiquinated proteins are recognized by the proteasome and degraded.

The proteasome is made up of a 20S cylindrical core containing the enzymatic activity, capped on both ends by 19S regulator complexes, each made up of a base and a lid, connected by another subunit. The base of the regulator complex consists of 6 ATPases and three other subunits, they are thought to recognises ubiquitin-tagged proteins and unfold and feed them into the 20S core; the function of the 9 unique non-ATPase subunits of the lid is unknown. The role of the whole 19S regulatory subunit is to recognize substrate, unfold the protein, and translate the substrate; it has a further role of removing the ubiquitin, as it needs to be recycled. Deubiquitilation is thought to be carried out by Rpn11.

The 6 ATPase subunits of the base of the 19S particle is arranged in a spiral hexameric ring, which resembles that of ClpX and helicases; thus its mechanism of protein translocation may be similar.

The 20S core is made up of four stacked rings, each ring contains 7 related proteins. The proteins are split into two groups (α and β) based on their sequence similarity. The outer rings are made up of the 7 α subunits (α1-α7), and the middle two rings are made up of the 7 β subunits (β1-β7).

The α-rings have two gates, one formed by the N-termini of the α-subunit that only opens when it is bound to the 19S subunit, to ensure that only proteins recognized by the 19S subunit enters the chamber to be cleaved; the second gate is 13Å-wide and this ensures that only unfolded proteins are able to enter into the chamber.

Three of the β subunits - β1, β2 and β5 - each have two active sites each to which it can perform its proteolytic activity; they seem to have a preference to cleave after acidic, basic or hydrophobic residues, respectively. They have activities similar to caspase, trypsin and chymotrypsin.

p97
p97 (a.k.a. VCP) is a Mg2+-dependent ATPase of the AAA family required for Golgi stack formations, nuclear envelop reassembly and tER formation. Because of its essential function, it is found relatively abundantly (1% of cytosol) and in all types of cells - CDC48 in yeast and VAT in archaea. Typical of AAA ATPases, its native conformation is a homohexamer.

It promotes Golgi stack formation by mediating membrane fusion. After from this, p97 functions in many pathways, including the Endoplasmic Reticulum Associated Degradation (ERAD) pathway, where it is used to extract misfolded proteins from the ER. It translocate the misfolded protein back the other way (towards the cytosol) and chaperon it to the proteasome for degradation. A similar mechanism also exists in the mitochondria and uses p97. p97 is also used for segregating protein complexes such as syntaxin V. Question arise as to how a single protein can serve so many functions. p97 consists of 3 domains - N, D1 and D2 - with D1 and D2 being the AAA domains. The two domains exists in different states (open/closed). In complex disassembly, motion in the N-domain is transferred to the substrate via an adaptor; major adaptors include p46 and Ufd1/Npl4.

When ATP is bound, it binds to the ubiquitin-target protein, and when ATP is hydrolyzed, it pulls the protein assembly apart.

AAA ATPase as molecular switches
AAA ATPase do not always couple ATP hydrolysis to mechanical force, some can also act as molecular switches. For the bacterial RNAP to function, an AAA ATPase must bind remotely to the promoter site. After binding to the DNA, it binds an ATP which allows it to associate with RNAP; after hydrolysis of the ATP, RNAP is now enabled to function. bEBP destabilized the DNA to open it up for RNAP to work on, as well as sliding the DNA into teh active site of the RNAP.

ATP hydrolysis is in some way affected by substrate (substrate it is translocating) binding, but the reverse is also true. Thus the AAA ATPase can act as a switch, switching between different affinities for the substrate.